UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prospective memory, a distinct form of remembering? : subtitle evidence from task comparisons and normal… Birt, Angela R. 2001

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

Item Metadata

Download

Media
831-ubc_2001-714381.pdf [ 12.53MB ]
Metadata
JSON: 831-1.0090713.json
JSON-LD: 831-1.0090713-ld.json
RDF/XML (Pretty): 831-1.0090713-rdf.xml
RDF/JSON: 831-1.0090713-rdf.json
Turtle: 831-1.0090713-turtle.txt
N-Triples: 831-1.0090713-rdf-ntriples.txt
Original Record: 831-1.0090713-source.json
Full Text
831-1.0090713-fulltext.txt
Citation
831-1.0090713.ris

Full Text

PROSPECTIVE MEMORY: A DISTINCT FORM OF REMEMBERING? EVIDENCE FROM TASK COMPARISONS AND NORMAL AGING by ANGELA R. BIRT B.A. (Hons), The University of Prince Edward Island, 1994 M.A., The University of British Columbia, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE R E Q U I R E M E N T S FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in THE FACULTY OF G R A D U A T E STUDIES (Department Of Psychology; Cognitive area) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 2001 © Angela R. Birt, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date 2.1c J&UJoXesynbtJu , ^OOf II ABSTRACT Until recently, memory research focused primarily on memory for information and experiences from the past (retrospective memory). However, memory also is important for remembering to do things in the future (prospective memory). Research on prospective memory is relatively new and has yet to address many pertinent issues. This dissertation had two main objectives: (1) examine core similarities and differences between prospective and retrospective memory and test the claim that prospective remembering is functionally distinct, and (2) investigate age differences in prospective memory test performance and their relation to age differences in performance on different types of retrospective memory tests. The nature of the relationship between prospective and retrospective memory was assessed with two quantitative literature reviews. Analysis of 148 prospective-retrospective memory task comparisons revealed many similar effects. Nevertheless, prospective memory test performance was more closely related to performance on some episodic retrospective memory tests than others. A meta-analysis of 96 young-old age difference effect sizes revealed age-related prospective memory declines comparable to those observed with episodic retrospective tasks, but the magnitude of age differences varied with task characteristics. Both quantitative reviews highlighted the large variability in the research. Next, a controlled laboratory experiment compared prospective and retrospective memory performance of 66 older (mean=75 yrs.) and 66 younger (mean=20 yrs.) adults. Matching prospective and retrospective test properties (e.g., cues, response type, context) enabled, for the first time, direct comparison of, and closer examination of age-related differences in, prospective memory tasks, retrospective memory tasks that decline with age (explicit cued recall and recognition), and a retrospective task that shows little decline (implicit word completion). Results indicated episodic prospective memory was not related to implicit memory, but was positively related to the tests of explicit memory. Self-rated prospective memory and individual differences in performance also were explored. Findings are discussed in relation to theoretical views of prospective memory and of aging. Although the results indicate that prospective memory is not a distinct form of memory, it appears to be a special application of explicit, episodic memory retrieval and is unique in the same way that free recall, cued recall, and recognition are unique activities of memory. iv TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES viii ACKNOWLEDGEMENTS x INTRODUCTION 1 CHAPTER ONE. Defining Prospective Memory 7 Defining Prospective Memory 8 Memory for intention versus memory for content 15 Tasks versus task components 17 Types of Prospective Memory Tasks 22 (1) Type of retrieval cue: Event- versus time-based tasks 22 (2) Retrieval cue frequency: Habitual versus episodic tasks 25 (3) Retention interval: Vigilance, monitoring, and prospective memory . . 27 CHAPTER TWO. Measuring Prospective Memory 30 Methods of Studying Prospective Memory 31 (1) Questionnaire and diary studies 32 (2) Naturalistic field studies 34 (3) Artificial laboratory studies 35 (4) Naturalistic laboratory studies 37 Single-intention, repeated-intention, and multi-intention designs 38 Issues of Measurement 40 (1) Restrict the type of task 42 (2) Minimize the retrospective component 43 (3) Lenient scoring criterion 44 Definitions and Methodologies: A Summary 46 V CHAPTER THREE. What Evidence is Needed to Argue for Functional Independence? 49 The Relationship Between Prospective and Retrospective Remembering . . . 50 Criteria proposed for identifying distinct memory functions or systems . . 52 Transfer-Appropriate Processing 59 CHAPTER FOUR. Theoretical Accounts of Prospective Memory Retrieval . 63 The "Recognition-Recall" Model of Prospective Memory 64 The "Automatic Associative Activation" Model of Prospective Memory 65 A Functional Account of Prospective Memory 68 CHAPTER FIVE. Prospective and Retrospective Memory Task Performance Compared 71 Evidence for (In)Dependence from Demonstrations of Functional and Stochastic Dissociations 72 Evidence Concerning Distinct Neural Pathways 75 Functional Incompatibility and Prospective Memory 82 CHAPTER SIX. Aging and Prospective Memory: Is There a Developmental Dissociation Between Prospective and Retrospective Remembering? . . . . 85 Retrospective and Prospective Memory: The Effects of Age 86 Prospective Memory and Aging: A Meta-Analysis 90 CHAPTER SEVEN. Self Assessment of and Individual Differences in Prospective Memory Functioning 94 Self-Assessment of Prospective Memory Performance 95 Individual Differences in Prospective Memory Performance 98 CHAPTER EIGHT. The Research Study 103 Summary of Study Objectives 104 METHOD 109 Participants 109 Materials and Design 111 vi Assessment of cognitive functioning 111 Stimulus materials 112 Levels of processing manipulation: The rating and counting task 113 Retrospective memory tests 114 Prospective memory tests 116 Individual difference measures 118 Procedure 121 RESULTS 129 Scoring and Initial Analyses of Retrospective Memory Performance 129 Word completion 129 Cued recall 131 Word recognition 132 Scoring and Initial Analyses of Prospective Memory Performance: Test 1. . . 133 Prospective memory test scores 133 Scoring and Initial Analyses of Prospective Memory Performance: Test 2, The Reading Task . . . 135 Prospective memory test scores 135 Reading time and self-report measures 136 Prospective memory errors 138 Source of prospective memory errors 140 Initial Performance, Forgetting, and Recovery Scores for Prospective Memory Tests 1 and 2 141 Functional Dissociations? The Effects of Age 142 Functional Dissociations? The Effects of Levels of Processing 146 Semantic vs. nonsemantic study conditions 147 Studied vs. nonstudied words 148 Stochastic Dissociations 149 Retrieval Context and Prospective Memory Test Performance 150 Retrieval context versus practice effects 152 Self-Assessment of Prospective Memory Performance 153 Individual Differences in Prospective Memory Performance 155 RESULTS DISCUSSION 159 vii GENERAL DISCUSSION . 172 FOOTNOTES 185 REFERENCES 186 TABLES 221 APPENDICES 242 Appendix A. Summary of prospective and retrospective memory task dissociations in the prospective memory literature 242 Appendix B. List of target words used in the retrospective memory tasks 249 Appendix C. Informed consent sheet (student version) 252 Appendix D. Response booklet 257 Rating & counting task 258 Prospective memory test 1 instructions 259 Demographics and health questionnaire 264 White Bear Suppression Inventory 266 Scoring sheet for word completion, cued recall, & word recognition tests 269 Word completion task 270 Cued recall task 273 Word recognition task 274 Prospective memory test 2 instructions 275 Modified Visual Search and Attention Test 276 Scoring sheet for the reading task 281 Reading task 282 "Agnes" -- Scoring scheme for the comprehension questions 285 North American Adult Reading Test 286 Prospective Memory Questionnaire 289 Dissociative Experiences Scale 298 Appendix E. "Agnes" by Gregory Clark 301 Appendix F. Debriefing form (student version) 305 viii UST OF TABLES Table 1. Summary of the Number of Dissociations Between Prospective (ProM) and Retrospective (RetM) Memory Task Performance Demonstrated in the Research to Date 221 Table 2 Participant Demographic and Health Characteristics 222 Table 3 Main Experimental Design 225 Table 4 Tasks Completed by all Participants Listed in Order of Administration 226 Table 5 Proportions Correct (M and SD) for the Implicit and Explicit Memory Tasks as a Function of Study Task and Age Group 227 Table 6 Prospective Memory (ProM) Test Scores as a Function of Age 228 Table 7 Mean Ratings of Task Difficulty, Interest, Absorption, and Willingness to Comply and Comprehension Scores for Prospective Memory Task 2: The Reading Task 230 Table 8 Summary of the MANOVA Assessing the Effects of Age (Two Groups) on the Retrospective and Prospective Memory Tests 231 Table 9 Summary of the MANOVA Assessing the Effects of Age (Three Groups) on the Retrospective and Prospective Memory Tests 232 Table 10 Effect of Study on Prospective Memory Performance (M and SD) . . . 233 Table 11 Pattern of Correlations Between the Memory Tests: Partial Correlations Controlling for Age and Correlations for the Young and Old Age Groups Separately 234 ix Table 12 Individual Difference Measures as a Function of Age 235 Table 13 Correlations Between Individual Difference Measures and Prospective Memory Task Scores for the Young Age Group 236 Table 14 Correlations Between Individual Difference Measures and Prospective Memory Task Scores for the Old Age Group 237 Table 15 Correlations Between Individual Difference Measures and Retrospective Memory Task Scores for the Young Age Group 238 Table 16 Correlations Between Individual Difference Measures and Retrospective Memory Task Scores for the Old Age Group 239 Table 17 Partial Correlations Between Individual Difference Measures and Prospective Memory Task Scores, Controlling for Age 240 Table 18 Partial Correlations Between Individual Difference Measures and Retrospective Memory Task Scores, Controlling for Age 241 ACKNOWLEDGEMENTS I would like to extend my sincere appreciation and gratitude to my supervisor, Peter Graf, Ph.D., for all his guidance, support, and advice over the course of my graduate studies. He has contributed immeasurably to my intellectual and professional development. I also would like to thank my committee members and examiners for their support and insightful suggestions and comments on my research: Erich Eich, Ph.D., Geoff Hall, Ph.D., Darrin Lehman, Ph.D., Mark Schaller, Ph.D., and John C. Yuille, Ph.D. Further thanks are offered to Kenneth Rockwood, MD, MPA, F R C P C of the Geriatric Medicine Research Unit, QEII Health Sciences Centre in Halifax, NS for his assistance and support in the recruiting and testing of the older participants, and to Stephen Porter, Ph.D. of Dalhousie University, Halifax, NS for providing laboratory space and research resources, for many discussions about memory, and for his research collaboration over the past five years. Many thanks also go to a number of people who helped with the data collection and management process, including Sarah Bell, Naomi Doucette, Monty Keough, Julie Kirk, Mike Kivell, Alicia MacDonald, Jennifer Ojiegbe, and Wanda Wilson. My deepest appreciation and gratitude go to Professor Owen Sharkey for there are no words to express how much his mentorship has meant to me. He is truly a special person and by far the best teacher I have ever had (and I have had many excellent teachers!). Over the last number of years, he has very generously provided me with many opportunities to learn and to grow, and he continues to serve as a source of inspiration. I cannot thank him enough for all that he has done for me. I have been very fortunate to have received a tremendous amount of support and understanding from my family and friends. In particular, I would like to thank my friends Tracy Casavant, Sarah Cockell, Cheryl Gilbert-MacLeod, Suzanne Hayman, and Jody Manley for being there for me in so many ways. I also thank my parents, Raymond and Deanna, and my sisters, Lisa and Tara, for always providing me with a level of support, understanding, and love that only a family can. Extra special thanks goes to my mother who, as usual, went far beyond the duties of motherhood to selflessly lend a helping hand in every way she possibly could. This dissertation would not have been completed without her. 1 INTRODUCTION Prospective Memory: A Distinct Form of Remembering? Evidence From Task Comparisons and Normal Aging When people complain about having a poor memory, they typically do not refer to difficulties they have remembering the article they read in the newspaper yesterday or the French they learned in grade school. They are more likely to be concerned over a tendency to forget appointments, to relay messages, or to run errands such as mailing a letter on the way to work. Memory research has traditionally focused on the study of retrospective memory, or the recollection of information and events from the past. However, recently, it has been recognized that another crucial function of memory is remembering to do things in the future. Memory researchers have begun to acknowledge the importance of studying how it is we "remember to remember" (Schonfield & Stones, 1979), a topic which has been coined prospective memory (Meacham, 1982; Meacham & Leiman, 1975/1982; Meacham & Singer, 1977)., Remembering previously formed intentions or plans is not only a common activity of memory, it is also essential to normal functioning. We are all intimately aware of the consequences of failures of prospective memory, which can range from the trivial to the extremely serious (Meacham, 1988). Forgetting to pick up a loaf of bread on the way home from work may not be a big deal, but showing up late for a job interview or forgetting to take heart medication can have serious consequences. The sheer number of ways in which we try to prevent such memory failures attests to the vulnerability of prospective memory. We use appointment books, planners, post-it notes, memos, electronic schedulers, timers, alarms, and even ask others to remind us 2 to remember! Obviously, in many cases we do not place a lot of trust in our ability to "remember to remember." Regardless of its fallibility, we rely on prospective remembering in order to function independently in daily life to such an extent that if that ability were taken away, we would be left severely incapacitated (Lezak, 1995). Most descriptions of prospective memory, like the one above, include a comparison with retrospective memory (see Brandimonte, Einstein, & McDaniel, 1996). Implicit in such comparisons is the assumption that prospective memory may be functionally distinct from retrospective memory. Although, intuitively, prospective memory seems to be quite different from retrospective memory, the critical differences between these two types of memory tasks are still a matter of debate (Ellis & Kvavilashvili, 2000). Problems with defining and measuring prospective memory retrieval have been at the heart of this debate and have led to mixed research findings. A variety of methodologies and experimental tasks have been used to study prospective memory functioning, ranging from measurement of appointment-keeping behaviour (e.g., Martin, 1986) and medication adherence (e.g., Ellis, 1998) in real world settings to strictly-controlled computer tests in the laboratory (e.g., Einstein & McDaniel, 1990) and self-report questionnaires (e.g., Hannon, Adams, Harrington, Fires-Dias, & Gipson, 1995). With such a range of tests purporting to test prospective "memory" and the variety of definitions and conceptualizations of prospective memory functioning, it is understandable that the research findings have been less than consistent. Thus, before the issue of whether prospective memory retrieval reflects a distinct function of memory can be addressed, it is necessary to examine the different ways in which prospective memory has been defined and measured. This dissertation begins with an in-depth analysis of the various definitions proffered for prospective memory. In Chapter One, particular attention is given to 3 identifying the specific properties that appear to make prospective memory retrieval unique and different from retrospective retrieval. The problems associated with the term "prospective memory" are addressed and, through a task analysis, the importance of distinguishing between prospective memory tasks and prospective memory task components is underscored. Prospective memory tasks can be broken down into a number of task components or processes, one of which is prospective memory retrieval. Prospective memory retrieval is defined here as unprompted cue recognition. That is, once a plan or an intention is formed, it is up to the individual to recognize a cue(s) in his/her environment as a sign of that previously formed intention, without any external prompts or reminders and in the absence of being in a "retrieval" mode. Because the focus is not on retrieval, recognizing the prospective memory cue is experienced as intent (a state of anticipation), not as memory for a previous experience. The variety of ways in which prospective memory tasks can differ are discussed. After highlighting the complexity and muti-faceted nature of prospective memory and noting that prospective tasks can be classified according to many of the same categories as retrospective tasks (e.g., short-term, long-term, episodic, semantic, procedural), the focus of discussion is narrowed to prospective tasks of an explicit, episodic nature. In addition, although the term prospective memory is often used as an umbrella term for a variety of tasks and mental processes, to reduce confusion and limit the scope of investigation, it will be used here to refer primarily to explicit, episodic prospective memory tasks and the mental processes required to perform them. Despite the fact that prospective remembering is so pervasive in everyday activities and is a fundamental aspect of memory, its popularity as a topic of research has begun to grow only recently. The most commonly cited explanation for the paucity of empirical research investigating prospective memory prior to the last decade is the 4 difficulty in designing well-controlled prospective memory tasks suitable for laboratory testing. In the last several years, progress has been made in the testing and measuring of prospective memory performance, and a number of experimental paradigms are currently utilized. The challenges of devising prospective memory tasks are discussed in Chapter Two and an overview of the different approaches to studying prospective memory is provided. Although a number of claims have been made about the nature of the relationship between prospective and retrospective task performance, very little research has tested directly the similarities and differences between the two. Chapter Three explores the different kinds of evidence that would be required to argue confidently that prospective and retrospective memory retrieval are functionally distinct. The basic evidence required for establishing independence between performance on different memory tests is a matter of considerable debate. Therefore, the most commonly proposed criteria considered as necessary for the postulation of distinct memory functions (or processes, or systems) are reviewed and discussed. To help further elucidate the ways in which prospective and retrospective memory retrieval might be related and/or distinct, an overview of some of the theoretical models and frameworks for prospective memory task performance is provided. Specific attention is paid to the predictions about the relation between prospective and retrospective memory tasks that follow from each model, for example, the notion that prospective memory retrieval is spontaneous and automatic and should closely resemble implicit memory retrieval (see Chapter Four). Next, the different criteria for postulating distinct memory functions/ systems are put to the test. The most common criterion used as evidence for independence is the presence of dissociations (especially double dissociations). As a preliminary test of the 5 hypothesis that prospective and retrospective memory tasks are functionally distinct, a quantitative review of the research on prospective memory to date was conducted. Specifically, the results of all the available research studies that have included tests of both prospective and retrospective memory were combined to determine the number of dissociations (both functional and stochastic) that have been demonstrated between the two tests. The results are presented in Chapter Five, along with a discussion of findings in the prospective memory literature relating to the various other criteria proposed. Initial studies of prospective memory and normal aging suggested that prospective remembering was preserved in older adults (e.g. Einstein & McDaniel, 1990). Subsequent research demonstrated that prospective memory can be impaired by normal aging unless effective external cues are employed (see Maylor, 1995; 1996). Nevertheless, compared to what has been discovered about age-related changes in retrospective remembering, relatively little is known about the effects of age on prospective memory functioning. For the past decade, research interest in prospective memory has been increasing, especially in the context of aging. Probably one of the greatest factors in this growing interest was Craik's (1986) framework for understanding age deficits in memory tasks. He proposed that memory test performance shows age-related deficits to the extent that it relies on self-initiated processing and lacks environment support (e.g., cues and context). Based on this assumption he argued that age-related impairments should be larger in prospective memory ("remembering to remember") tasks than retrospective memory tasks due to greater demands on self-initiated processing (i.e., unprompted cue recognition). This proposal was investigated further in this dissertation. The extent to which unprompted cue recognition declines as a function of age was examined in two ways. First, a 6 comprehensive meta-analysis of the young-old age differences in prospective memory test performance within the extant research literature was conducted (see Birt, 1999), and second, a research study comparing the performance of young and older adults on tasks of prospective memory was conducted and the sizes of age-related performance differences were compared with those observed on a number of different retrospective memory tasks. The goal was to gain a better understanding of the effects of age on prospective remembering, and by testing for developmental dissociations to establish whether prospective and retrospective memory task performance are functionally distinct. Chapter Seven highlights the need for an increased understanding of individual differences in personality, cognitive abilities, and meta-cognitive processes for gaining insight into potential sources of the variability in prospective memory performance that most researchers have observed. The self-assessment of prospective memory abilities and how such self-evaluations may relate to actual memory performance are discussed. In addition, a variety of potential individual difference factors, such as verbal ability, visual search capacity, dissociative tendencies, and thought suppression are explored. The final chapter, Chapter Eight, is a report on the research study that was conducted to test the relationship between prospective and retrospective memory performance and to investigate potential age differences in prospective memory retrieval. The methodology and results are presented in detail and the implications of the findings are discussed, especially with respect to new insights gained into the nature of prospective memory and changes in memory accompanying normal aging. CHAPTER ONE Defining Prospective Memory 8 Defining Prospective Memory Some examples of prospective memory tasks include remembering to mail a letter on the way home from work, remembering to meet a friend at 2:00 p.m., remembering to wake up an hour earlier than usual in the morning, remembering to pass on a message to a colleague, and remembering to take medication with meals. What do these memory tasks have in common? How do such tasks differ from retrospective memory tasks? There are at least five different ways in which investigators of prospective memory have tried to capture the commonality or essence of different prospective memory tasks and distinguish them from tasks of retrospective memory. Prospective memory task performance has been defined in terms of: (1) the temporal aspects of the memory; (2) the spontaneous quality of remembering; (3) the to-be-remembered content; (4) various noncognitive task components involved; and (5) the specific task characteristics and demands. Each of these definitional approaches are discussed in turn. One of the more obvious ways of attempting to define prospective memory tasks and distinguish them from retrospective memory tasks is in terms of the temporal aspects or temporal quality of the memories themselves (e.g., Meacham, 1982). Although all memories are formed in the past, not all memories are necessarily about the past. According to Malcolm (1977), in order for a "memory" to be a memory (i.e., memory proper), its contents do not need to be about the past, but it must be accompanied by the knowledge or awareness that it was learned or experienced in the past (see also James, 1890 for an early discussion). Following this logic, it can be argued that the key distinction lies in the fact that retrospective memories are about the past, and prospective memories, although formed in the past, are about the future. 9 Support for this claim comes from phenomenological differences (Hunt & Smith, 1996). Prospective memories have a feeling of futurity and anticipation. By contrast, retrospective memories have a feeling of "pastness" and history. However, this distinction may not be so clear-cut. As Meacham (1982) points out, not all memories about the future are prospective memories. For example, one can remember that the weather is supposed to be warm and sunny tomorrow or that Christmas will be on a Tuesday this year, but even though these memories are about the future, they are not instances of prospective remembering. Therefore, although prospective and retrospective memories seem to be characterized by different temporal qualities (and these qualities do result in phenomenological differences), the defining distinction does not appear to be one of memories being about the future versus the past. A second approach to defining prospective memory focuses on the spontaneous quality of prospective remembering. Previously formed intentions often seem to "pop" to mind suddenly in response to a particular cue while one is engaged in other activities. Many researchers have emphasized the spontaneous nature of prospective remembering (e.g., Blackburn, 1995; Einstein & McDaniel, 1996; Mandler, 1994; Mantyla, 1996; McDaniel, 1995; Meacham, 1982), including Freud (1901) who wrote: If I form an intention in the morning which is to be carried out in the evening, I may be reminded of it two or three times in the course of the day. It need not however become conscious at all through the day. When the time for its execution draws near, it suddenly springs to mind and causes me to make necessary preparations for the proposed action (p. 152). 10 A distinction can be made between remembering that is directed, controlled and voluntary versus memory that is spontaneous, automatic, and involuntary (e.g., Ebbinghaus, 1885/1964; Mandler, 1994; Moscovitch, 1994; Ste-Marie & Jacoby, 1993; Tulving, 1983). Directed memory is purposeful, deliberate, and controlled and occurs as a result of a direct question or a direct prompt to recollect. Spontaneous memory is more automatic and involuntary and occurs as a result of a cue or cues in the environment triggering memory reflexively. Despite the fact that spontaneous remembering does appear to be one of the qualities of many instances of prospective remembering, it does not appear to capture the core difference(s) between prospective and retrospective memory retrieval (e.g., Crowder, 1996; Meacham, 1982). Although most of the laboratory tests of retrospective remembering involve direct prompting and cueing from the experimenter, many, indeed most, of our retrospective memories in everyday life are spontaneous and automatic. Retrospective memories often "spring" to mind as a result of either internal cues and/or external cues in the environment (e.g., Mandler, 1994; Ste-Marie & Jacoby, 1993). For example, one tries to remember a name and, being unsuccessful, abandons the search only to have the name suddenly pop to mind later in a seemingly unrelated context. Similarly, one struggles for a period of time to solve a difficult puzzle or problem, then, while taking a break, the solution unexpectedly "appears" as if out of nowhere. Thoughts, facts, perceptions, and ideas frequently come to mind unexpectedly without any deliberate attempt to remember them. In fact, a substantial amount of research has been dedicated to the study of the spontaneous nature of many cognitive processes. Examples include: incubation (nondeliberate, usually sudden occurrence of an idea or problem solution after a delay in retrieval attempts; e.g., Posner, 1973; Smith & Blankenship, 1991), mind popping (spontaneous, unintentional remembering while engaged in unrelated activities; e.g., 11 Mandler, 1994), and tip-of-the-tongue effects (temporary retrieval block or inaccessibility to a specific piece of information accompanied by memory for related information; e.g., Brown, 1991; Brown & McNeill, 1966). Thus, it appears that retrospective memory retrieval can occur as spontaneously and involuntarily as prospective memory retrieval. Similarly, prospective remembering can occur in a directed or controlled fashion, as is often the case when the intention is novel, when the retrieval context is poorly specified, and when external cues are implemented to aid remembering (Ellis, 1996). A third, and by far the most common, way of attempting to define prospective memory tasks has been to focus on the to-be-remembered content of the memory. Examples of such definitions include "memory for intentions" (Goschke & Kuhl, 1996; Kuhl & Kazen, 1999), "memory for future actions" (Einstein & McDaniel, 1996; Mantyla, 1993), "memory for plans and actions" (Cohen, 1989), "memory for an intended action" (Kvavilashvili, 1998), and "remembering that something has to be done" (Maylor, 1996b). According to these definitions, it is what has to be remembered that is of utmost importance-the intention, the action, the plan, the something that has to be done. Whereas episodic retrospective memory concerns memory for events, prospective memory deals with memory for intentions. The assumption here is that there is something inherently different about remembering an intention from remembering any other piece of information. Exactly what that difference might be is not clearly specified. Such content-specific definitions are descriptive; they do not explain what is unique about prospective memory tasks or add to our understanding about intention or memory (Craik & Kerr, 1996; Graf & Uttl, 1999; Rabbitt, 1996). "Intentions, like other mental events, can be remembered or forgotten afterwards" 12 (Crowder, 1996, p. 146). How is remembering a previously formed intention different from remembering anything else from the past? Is using memory to plan future behaviour different from using it to reconstruct the past? It is definitions of this type that have spurred much of the criticism of and debate within prospective memory research, leading some memory researchers to argue that the term prospective memory is misleading and even careless (e.g., Crowder, 1996; Roediger, 1996). A fourth way in which prospective memory tasks have been defined and considered to be unique is according to what Winograd (1988) referred to as the noncognitive components important for prospective memory performance (e.g., Dobbs & Reeves, 1996; Einstein & McDaniel, 1996; Meacham, 1982, 1988; Park & Kidder, 1996). Compared to retrospective remembering, prospective remembering is viewed as being more concerned with non-mnestic processes; that is, activities and behavioural factors not usually regarded as being about memory (Burgess & Shallice, 1997; Dobbs & Reeves, 1996; Shimamura, 1990; Shimamura, Janowsky, & Squire, 1991; Winograd, 1988). According to this view, the study of prospective memory is more concerned about "real world" characteristics of behaviour and action such as attention, metaknowledge, goal-setting, planning, problem-solving, motivation, compliance, monitoring, vigilance, reward, interpersonal relations, output evaluation, reality monitoring, etc. than about memory per se. Prospective memory task performance is considered multidimensional, with each of the above factors/ components making a unique contribution. There is little doubt that all of these factors play an important role in determining prospective memory performance. However, even though the claim has been made (e.g., Winograd, 1988), it is not clear that they play any less of a role in retrospective memory task performance. Attention, motivation, monitoring, metaknowledge, etc. all influence the effectiveness of 13 retrospective recollection and, therefore, do not provide any clarification as to the critical distinction between retrospective and prospective memory retrieval. Furthermore, many of these noncognitive factors (e.g., attention, motivation, metaknowledge) have more to do with subject characteristics than required components of prospective memory tasks themselves. A fifth way of defining prospective memory tasks is in terms of the specific task characteristics and requirements (e.g., Craik & Kerr, 1996; Einstein & McDaniel, 1996; Graf & Uttl, 1999). Any memory task can be examined objectively according to the cues, the instructions, the context, and the response type (e.g., Graf & Birt, 1996), and by comparing different memory tasks according to such attributes, important distinctions can be revealed. Probably the best way to try to isolate the critical differences between any two memory tasks is to equate the tasks on as many dimensions as possible (e.g., Graf & Mandler, 1984; Schacter, Bowers, & Booker, 1989). If this approach were applied to explicit, episodic prospective and retrospective memory tasks, it would be possible, without losing the essence of the tasks, to use the same cues at test, the same context, and even require the same type of response, but the specific test instructions would have to differ. That is, it appears to be the instructions concerning the use of the test cues that differ critically between the tests. As noted by Graf and Uttl (1999), a task comparison reveals that although cues are provided at the time of test for both memory tasks, what differs is: (1) whether the subjects are alerted to the cues and asked to work with them in some manner; and (2) whether they are informed of the connection between the test cues and the previous study/ planning phase of the experiment. In contrast to explicit retrospective memory tasks, during prospective memory task performance participants are not made aware 14 of the presence of the cues and they are not explicitly asked to work with the cues at test. Experimental tests of prospective memory require that subjects perform a "cover" or ongoing task, and the prospective cues appear as a normal part of the activities involved in that ongoing task. It is up to the subjects to notice and recognize the cue(s), that appears naturally as part of ongoing thoughts or activities as a reminder of their previously formed intention or plan, without being prompted to do so (Craik, 1986; Graf & Uttl, 1999; Maylor, 1996a, 1996b; Meacham, 1982; Winograd, 1988). Thus, subjects must first recognize the cue (which may not stand out or be very distinct within the context in which it is embedded) as a cue for an intention before using it to retrieve the intended action or plan. It is this unprompted cue recognition that makes prospective remembering different from retrospective remembering. Although the mechanisms for actually noticing or recognizing a cue in retrospective memory retrieval and prospective memory retrieval may be essentially the same, the subjective experience in the former will be of remembering, not intent (see Hunt & Smith, 1996). In other words, because subjects are not alerted to the connection between the retrieval cue and the previous experience of encoding or forming the intention, they are not in a "retrieval" mode (Tulving, 1983) when unprompted cue recognition occurs. In fact, attention is almost always directed elsewhere—at ongoing thoughts and activities. What pops back to conscious awareness when the retrieval cue is successfully recognized among these ongoing activities is an intention, not a previous experience. Once the intention to do something is remembered, this may then trigger or induce a "retrieval" mode and initiate a deliberate memory search for exactly what the previously formed intention was. 15 Memory for intention versus memory for content. The decomposition of prospective memory tasks into their component processes has been carried out to varying degrees of specificity and a variety of these components have been highlighted as critical for determining task performance. Nevertheless, one distinction between component processes has been emphasized by most prospective memory researchers: the distinction between memory for an intention and memory for the content of that intention (e.g., Einstein & McDaniel, 1990, 1996; Ellis, 1996; Kvavilashvili, 1987; Maylor, 1996a, 1996b; Winograd, 1988). Consider the example of wanting to stop at the grocery store on the way home from work. It is possible to remember: (a) both the intention (stopping somewhere on the way home) and the content (the grocery store); (b) the intention but not the content (remembering wanting to make a stop, but not remembering where); (c) the content but not the intention (dinner preparations trigger the realization that the stop at the grocery store was forgotten); and (d) neither the intention nor the content (no stop at the grocery store was made and there is no realization that the stop was forgotten). Failures to carry out a previously formed intention can happen for any number of reasons, including reasons pertaining to problems with memory per se and reasons relating to the aforementioned noncognitive factors, such as motivation and attention. Given that the many of these noncognitive factors are subject-related factors and most can be controlled in the experimental laboratory situation, the primary task for memory researchers is to determine whether a memory failure has occurred and identify its source. In order to do this, it is important to distinguish between remembering that something has to be done (the intention) and remembering exactly what it is that has to be done and when and where (the content). 16 In an attempt to gain a better understanding of prospective memory retrieval, it is interesting to consider further the case of successfully remembering an intention followed by an unsuccessful attempt at remembering the content of that intention. If the retrieval cue is successful in triggering memory for the intention and a search for the content of the intention is initiated, but the content of the intention is temporarily inaccessible, what is the quality of the resultant experience? Memory failures of this kind happen quite frequently, especially with the elderly. Examples include forgetting what you went into a room to get, beginning to do something and then forgetting what it was you wanted to do, and forgetting what you wanted to say in the middle of a sentence. As discussed earlier, such temporary retrieval blocks appear to be very similar to those experienced with the tip-of-the-tongue effect (Brown, 1991; Brown & McNeill, 1966). In both cases, the remember is in a retrieval mode attempting to remember specific information. Despite conscious, deliberate attempts at retrieval, a feeling of familiarity or retrieval fluency, and often-successful retrieval of related information, the desired information remains inaccessible to conscious awareness. The only difference between tip-of-the-tongue experiences that occur in the context of prospective memory tasks versus those that occur in the context of retrospective memory tasks is that memory for an intention to act serves as the trigger in the former and memory for a previous experience or piece of information serves as the trigger in the latter. Einstein and McDaniel (1990; 1996) refer to the intention and the content of the intention as the prospective and the retrospective components of prospective memory tasks, respectively. However, as discussed earlier, it is not simply the memory for the intention (as opposed to its content) that makes a prospective memory task unique. In fact, it is well known in the memory literature that the content of an episode can be 17 dissociated from the memory for the episode itself. For example, I may remember that I parked my car this morning, but I may not be able to remember where I parked it. Similar examples include recalling being introduced to someone, but being unable to remember the individual's name, and remembering the act of writing down a phone message, but not recalling any portion of it. Therefore, the distinction between remembering that something is to be (or has been) done and the specific contents of that memory is not a defining feature of prospective memory tasks in and of itself. It can apply to both prospective and explicit retrospective tasks. The unique part of the prospective task involves recognizing a cue as an indication of a previously formed intention in the absence of any prompting to do so and in the absence of being in a "retrieval" mode. After the intention is brought back to awareness, the task is then guided by retrospective memory retrieval in order to retrieve the content of the intention. Tasks versus task components. The distinction between memory tasks and memory task components is of utmost importance and has been a source of much of the confusion in defining and evaluating prospective memory performance. The prospective memory literature is highly variable in terms of the focus and scope of inquiry. Researchers who are interested in overall task performance have argued that prospective memory performance is "more than memory" (Dobbs & Reeves, 1996), is more concerned with various "noncognitive components" (Winograd, 1998), and should be renamed the "realization of delayed intentions" (Ellis, 1996; Kvavilashvili & Ellis, 1996). Such arguments seem warranted because, like retrospective memory tasks, the broad domain of prospective memory task performance can be divided into many different task components (the most general of which are encoding, storage, and 18 retrieval) and can be influenced by many different subject- and task-related variables (e.g., motivation, attention, reward, etc.). However, it has been argued that the term prospective memory, as currently conceptualized, is too broad to be useful (e.g., Cohen, 1989; see Ellis & Kvavilashvili, 2000; cf. Burgess, 2000). Recognizing this problem, other researchers (e.g., Graf & Uttl, 1999, Einstein & McDaniel, 1990; 1996; Einstein, Holland, McDaniel, & Guynn, 1992; Kvavilashvili, 1998; Mantyla & Nilsson, 1997; Maylor, 1996b, Uttl, Graf, Miller, Mclsaac, & Tuokko, 1999) have narrowed their investigations to prospective memory retrieval, which is only one task component of prospective memory tasks. This level of inquiry is much more focused and is concerned primarily with memory functioning per se. A close examination of various prospective memory tasks indicates that successful performance can be broken down into several general subtasks, task components, or performance phases (e.g., Brandimonte & Passolunghi, 1994; Cockburn & Smith, 1994; Dobbs & Reeves, 1996; Ellis, 1996; Harris, 1984; Hitch & Ferguson, 1991; Meacham, 1982; Norman, 1981; 1988). These phases include: A. Forming an overall goal-deciding on a desired outcome B. Forming an intention to do something in the future to achieve the goal C. Exploring the possibilities of what to do D. Deciding or choosing what to do E. Creating a plan of action-identifying, choosing, and organizing the steps required to achieve the intended action F. Generating or selecting a retrieval cue(s) that will serve as a trigger to remember the intention at the appropriate time G. Retention interval 19 H. Recognizing the cue as a reminder of the previously formed intention-remembering that something has to be done I. Initiating recall of the contents of the intention and plan of action J . Completing (or abandoning) the intended action K. Evaluating the outcome with respect to the initial goal Phases A through F are concerned with the encoding of the intention. Once an overall goal has been established and an intention to do something has been formed, the possibilities of potential actions are explored and weighed according to both previous experience and current motivations and situational demands. From the various possibilities, a to-be-performed action is chosen and a plan is constructed outlining the steps necessary to carry out that action. To ensure that the intended action is performed at the appropriate time, a retrieval cue(s) (which can range from very general to quite specific) is selected to serve as a sign or reminder of the previously formed intention. According to Ellis (1996, 1998), the intention indicates that you have decided to do something and the action refers to what it is you intend to do. Phase G refers to the amount of time that passes between encoding and retrieval of the intended action. It is important to note that, as is the case with retrospective memory tasks, the size of this interval can range from being quite short to quite long. For example, waiting for a traffic light to turn green in order to proceed through an intersection is a relatively short-term task and is likely to remain active in working memory and dominate conscious awareness. By contrast, the task of remembering to stop at the grocery store on the way home from work is unlikely to remain active and dominant in working memory; it is most likely out of conscious awareness for all (or most of) the retention interval. In this case, because the retention 20 interval is longer it is filled with other activities (e.g., driving home from work). The prospective cue (e.g., the store) is embedded in the context of these activities which makes it more difficult to be recognized as a sign of the previously formed intention than in the case of the short-term prospective task. The bulk of prospective memory research has been concerned with tasks employing retention intervals that exceed short-term memory capacity, and unless otherwise specified, the prospective memory tasks referred to in this dissertation will be of this type. Inhibition also has been noted as important during the retention interval phase of the task. Between the time the intention is formed and the appropriate performance context manifests, it is necessary to suppress or inhibit the premature enactment of the intended action (e.g., Kuhl & Kazen, 1999). However, this inhibitory process is likely more important and more challenging to maintain in prospective memory tasks with short retention intervals (e.g., vigilance, monitoring) in which the retrieval cue could occur at any moment. Phases H and I concern the retrieval of the intention. In contrast to retrospective memory tasks in which recall is typically directly prompted, memory retrieval in prospective memory tasks occurs in the absence of any explicit prompting or instruction for remembering. As stated previously, it is up to the individual performing the task to identify or recognize the cues as signs or reminders (of previously formed intentions) when they appear embedded in and as a natural part of other ongoing activities, thoughts, or situations. Once the cue has been recognized as a sign that something is to be done (i.e., once the intention to act is remembered), a deliberate search of memory must be initiated to think back to the previous experience of forming the intention and recall the content of that intention (i.e., exactly what action was intended and the plan that was devised to carry out that action) (e.g., Einstein & McDaniel, 1996; Kvavilashvili, 1987; Maylor, 1996b). In phase J , the task is actually 21 carried out. However, it is important to note that the person, for any number of reasons (motivation, compliance, current situational demands, etc.), may choose not to carry out the intended action after all. The final phase, phase K, concerns evaluating the degree to which the intention has been satisfied and forming a memory for the outcome. This performance evaluation helps to prevent repeating an intended action that has already been carried out or failing to perform one that has yet to be carried out. Reality monitoring (e.g., Johnson & Raye, 1981) and output monitoring (e.g., Koriat, Ben-Zur, & Sheffer, 1988) have been identified as important for this evaluation phase. For example, reality monitoring can help with distinguishing between the memory of an intention to perform an action and the memory for the experience of performing that action, preventing errors of omission, and output monitoring can aid in remembering whether or not a planned action has already been performed thereby circumventing errors of commission (Maylor, 1996b). To summarize, prospective memory task performance can be divided and analyzed according to a number of performance components or phases. These task components can be broadly categorized into a general encoding phase (phases A through F), a storage phase (phase G), a retrieval phase (phases H and I), and execution and evaluation phases (phases J and K). The distinction between memory tasks and memory task components has important implications for not only defining and clarifying the scope of inquiry into prospective memory, but also for measuring and evaluating prospective memory performance. It is obvious that unless steps are taken to try to isolate the retrieval phase/ component, measuring overall performance in prospective memory tasks may not be especially informative about prospective memory retrieval per se. 22 Types of Prospective Memory Tasks Consider the following examples of prospective memory tasks: (a) remembering to pass on a telephone message to a friend, (b) remembering to return a library book before its due date, (c) remembering to take your purse or wallet with you wherever you go; (d) remembering to meet a friend in 20 minutes, and (e) remembering to take medication with meals. As discussed previously, all prospective memory tasks involve forming an intention and later recognizing naturally occurring cues as reminders of that previously formed intention. Nevertheless, it is obvious from the examples above that prospective tasks can vary substantially with respect to many important task characteristics. Retrospective memory has long been recognized as including different types of activities requiring functions or systems that are phenomenologically distinct (e.g., short- vs. long-term, episodic vs. semantic, explicit vs. implicit, declarative vs. procedural). Therefore it is possible that distinct forms or activities of prospective memory also can be identified. Prospective memory researchers have distinguished between types of prospective memory retrieval tasks according to a number of task features or dimensions. Several of these dimensions include: (1) the type of retrieval cue, (2) the frequency of the retrieval cue presentation, and (3) the retention interval. Each of these will be discussed in turn. Examining the variety of ways in which prospective memory tasks can differ and attempting to identify the key similarities of such tasks is important in order to focus and clarify discussion as well as to understand the pattern of findings (especially the inconsistencies) in the existing literature. (1) Type of retrieval cue: Event- versus time-based tasks. One task feature that has been identified as important for differentiating prospective memory tasks is the 23 type of retrieval cue or context. Retrieval cues are critical for prospective memory performance as they serve as a reminder that something needs to be done. Some possible prospective memory cues include: activities, locations, persons, objects, events, times, or time periods (Harris & Wilkins, 1982; Kvavilashvili & Ellis, 1996). Do different types of cues have different effects on prospective memory task performance? Harris (1984) acknowledged a difference between prospective memory cues when he distinguished between tasks that require "remembering to keep appointments" versus those that require "remembering to do one thing before or after another." Drawing from this distinction, Einstein and McDaniel (1990; 1996; McDaniel & Einstein, 1992) categorized prospective memory tasks into two general types-event-based and time-based tasks-based on the type of retrieval cue employed in the task. Event-based tasks require remembering a previously formed intention after some event occurs. The prospective memory cue is an external event (e.g., a mailbox) in the case that noticing it triggers memory for an intention (e.g., mailing a letter). By contrast, time-based tasks require remembering a previously formed intention at a specific time or after a certain period of time has elapsed. In this case, the prospective memory cue is time, for example, remembering to attend a meeting at 2:00 p.m. or turning off the microwave in two minutes. Einstein and McDaniel (1990, 1996) argue that what makes event- and time-based tasks fundamentally different is that event-based tasks have external cues that serve to initiate and guide memory, whereas time-based tasks do not (or at least a less salient cue). Therefore, time-based cues are assumed to provide less environmental support and necessitate more self-initiated monitoring and memory retrieval processes than event-based cues (see Craik, 1986). However, Rabbitt (1996) offered an alternative explanation for differences between time- and event-based prospective 24 memory tasks. He argued that time-based tasks are more demanding because the monitoring of the passage of time necessary for successful performance is a secondary task that can interfere with remembering previous intentions or plans (see also Ceci , Baker, & Bronfenbrenner, 1988; Harris & Wilkins, 1982). To illustrate, if you plan to attend a meeting in 20 minutes, then you must keep track of the time as you perform other activities. On the other hand, if you decide to set an alarm to ring in 20 minutes, then you can continue with your activities without having to think about or check the time at all. Kvavilashvili and colleagues (1990; Kvavilashvili & Ellis, 1996) expanded upon the event- versus time-based task dichotomy by proposing a third basic type of prospective memory task involving yet another class of retrieval cues-activity-based tasks. Kvavilashvili defines event-based cues as those that are relatively independent of a particular individual and activity-based cues as stemming from the activities that a particular individual performs. Activity-based tasks require that the individual identifies his or her own actions as a cue (e.g., remembering to set the alarm before going to sleep), whereas event-based tasks require the identification of a cue which is independent of or external to one's own actions. Regardless of the specific categories used to classify prospective memory tasks, it seems apparent that prospective memory tasks can differ dramatically according to the specific retrieval cue. In fact, prospective memory task performance has been found to differ depending on the type of cue utilized. For example, relative to younger adults, older adults tend to show significantly poorer performance with time cues than event cues in laboratory-based tests of prospective memory (see Birt, 1999). It is also important to note that retrieval cues/ contexts are usually much more complex than just an event, time, or activity, especially in the real world (Burgess & Shallice, 25 1997). Consider the example of remembering to set the alarm clock before going to sleep. This retrieval context includes a time (bedtime), an activity (going to sleep), a location (bedroom), and an object (the alarm clock). It is possible that any one of these retrieval cues could be utilized to remember the intention to set the alarm. Nevertheless, it is all of these cues together that form the rich, complex retrieval context, which will hopefully facilitate remembering. (2) Retrieval cue frequency: Habitual versus episodic tasks. Prospective memory tasks can also differ according to the frequency with which they are performed. Some intentions are carried out frequently on a regular, routine basis-for example, remembering to bring your purse or wallet wherever you go or taking a coffee break each morning at 10:00 a.m. Other intentions are executed irregularly or infrequently (remembering to schedule an eye exam) or are carried out (hopefully) only once (remembering to show up for your Ph.D. defense). Meacham and Leiman (1975/1982) referred to these as habitual and episodic prospective memory tasks, respectively (see also Harris, 1984). Meacham and Leiman (1975/1982) suggested that habitual intentions are easier to remember because they are often embedded in the context of a series of activities, routinely performed in a certain order. This rich contextual environment and ordered sequence of activity provides the necessary cues for remembering what to do next. Completing one activity leads to the initiation of the next, often automatically. Indeed, in general, as the frequency in which an action (physical or mental) is performed increases, the more habitual or automatic the performance becomes (e.g., Hasher & Zacks, 1979). Thus, the more often an intention is formed and subsequently remembered, the more automatic its execution becomes (Ouellette & Wood, 1998). 26 For example, if I began buying a coffee every morning on my way to work, this behaviour would soon become habitual and I would become less aware of and would have to devote fewer cognitive resources to my intention to do so every morning. This is possible because a specific stimulus (or stimuli) can come to serve as a trigger to activate routine action schemas or automatized skills with little or no conscious experience of an anticipation or intention (e.g., Norman & Shallice, 1986; Searle, 1983). By contrast, episodic intentions require one to break the normal flow of daily activities, often interrupting habitual, routine behaviours to perform a novel action. Because episodic intentions are irregular, relatively novel, and not embedded in the context of routine activities, there are fewer retrieval cues available to trigger memory for them (e.g., preparing and packing for a vacation). Therefore, in general, it appears that episodic prospective memory tasks should rely on effortful and controlled processing to a greater extent than habitual prospective tasks. However, it is recognized that not all mental processes/ behaviours necessarily need to be well-practiced in order to become habitual or automatic and retrieval cues and contextual information can be processed relatively automatically, resulting in fluent prospective memory retrieval (see Gollwitzer, 1999; Gollwitzer & Brandstaetter, 1997; Jacoby, 1991; Ste-Marie & Jacoby, 1993). A more detailed analysis of episodic intentions reveals that they can be further divided into single and repeated intentions (Kvavilashvili, 1992). A single episodic intention is one that need only be remembered or carried out once in response to a single occurrence (or recognition) of the retrieval cue or context. A repeated episodic intention is one that has to be remembered on several different occasions in response to multiple occurrences of the retrieval cue/ context. Similar to habitual intentions, repeated intention tasks do not require that a new intention be formed on each 27 occasion, but, unlike habitual intentions, single intentions are not performed routinely or on a regular basis. That is, although the same intended action is remembered or carried out more than once, it is not performed often enough to be considered habitual or automatic. The importance of repeated intention designs in testing and measuring prospective memory performance is discussed in greater detail in a subsequent section addressing issues of methodology. (3) Retention interval: Vigilance, monitoring, and prospective memory. A third task feature that can vary between different prospective memory tasks is the length of the retention interval (i.e., the period of time between forming and remembering an intention). Retrospective memory performance has been investigated as a function of a wide variety of retention intervals, ranging from a few seconds, to minutes, days, months, and even years. One area of retrospective memory research that has received a plethora of attention is the distinction between short-term and long-term memory tasks. Different processes and systems have been postulated to account for performance differences between these memory tasks. Baddeley and Wilkins (1984) proposed that a similar distinction can be applied to prospective memory. They argued that successful performance on many shorter-term prospective memory tasks may require holding the intention active in "conscious awareness" during the retention interval. In addition, Meacham and Leiman (1975/1982) suggested that short-term prospective tasks may be no different from tasks of vigilance for which one's attention must be maintained on the goal of the task throughout its duration. Thus, it can be argued that even tasks with little or no retention interval, such as vigilance and monitoring, can be considered types or forms of prospective memory tasks (Baddeley & Wilkins, 1984; Burgess, 2000; Graf & Uttl, 1999; Meacham & Leiman, 1975/1982). 28 Vigilance, monitoring, and prospective memory tasks can be viewed as part of a continuum according to the degree to which the intention occupies conscious awareness, with vigilance at one end, prospective memory tasks at the opposite end, and monitoring somewhere near the middle (Craik & Kerr, 1996; Graf & Uttl, 1999). However, most definitions of prospective memory and intentions make reference to an interval of time in which no thought was given to the intended action or plan. Freud (1901) defined an intention as "an impulse to perform an action: an impulse which has already found approval but where execution is postponed to a suitable occasion" (p. 201). It has been recommended that the term prospective memory be reserved for the case in which there is a significant period of time during which the previously formed intention does not enter (or rarely enters) conscious thought (e.g., Graf & Uttl, 1999; Maylor, 1996). Consistent with this recommendation and unless otherwise specified, the term prospective memory will be used here to refer to prospective memory tasks of an episodic nature with retention intervals that exceed short-term memory capacity. Many other distinctions between types of prospective memory tests have been recommended based on various task features, including single- versus dual-activity tasks (Harris, 1984), simple versus complex tasks (Einstein, et al., 1992; Rabbitt, 1996), self- versus other-generated intention tasks (Cohen, 1989), important versus unimportant intention tasks (Cicogna & Nigra, 1998; Meacham & Singer, 1977), etc. It is predictable that, as is the case with retrospective memory, some of these task properties will be more critical in determining prospective memory task performance than others. Some may be important during the encoding phase in which goals are formed and plans are made, whereas others may have more influence on cue recognition during the retrieval phase. Nevertheless, it is important to note that the features of prospective memory tasks can vary greatly, and it is possible that some of 29 these features may be key in identifying functionally and phenomenologically distinct forms of prospective remembering. CHAPTER TWO Measuring Prospective Memory 31 Methods of Studying Prospective Memory Designing methods for studying prospective memory presents a challenge. At the time of this writing, there have been only a little over a hundred empirical research papers devoted to prospective memory. Probably the most significant factor accounting for the small number of research studies is the difficulty in designing well controlled tasks that provide pure measures of prospective memory retrieval (Dobbs & Reeves, 1996; Kvavilashvili, 1992, 1998; Maylor, 1996b). One purpose of the present research was to attempt to refine current methods for testing and measuring prospective memory retrieval. This chapter gives a comprehensive overview of the various methodologies used to assess prospective memory to date, outlining the strengths and weaknesses of each. In any prospective memory task, one has to be sure that an intention is actually formed and that a retrieval cue is set up and encoded sufficiently. This information must then be retained over a period of time during which the subject is distracted and presumably not thinking about the prospective task. The subject must notice and respond to the cue when it is presented, without being prompted to do so, in such a way that indicates that he or she remembers the previously formed intention (i.e., they at least remember that something has to be done). 1 As many researchers have discovered, creating an experimental paradigm to test prospective memory performance is not an easy task. Most existing research on prospective memory can be classified into four general experimental methodologies: (1) questionnaire and diary studies, (2) naturalistic field studies, (3) artificial laboratory experiments, and (4) naturalistic laboratory studies. 32 (1) Questionnaire and diary studies. A number of questionnaire studies have investigated self-reports of prospective memory abilities (e.g., Andrzejewski, Moore, Corvette, & Herrmann, 1991; Bennett-Levy & Powell, 1980; Cohen & Faulkner, 1984; Dobbs & Rule, 1987; Hannon et al., 1995; Herrmann & Neisser, 1978; Hertzog, Park, Morrell, & Martin, 2000; Martin, 1986; Mateer, Solhberg, & Crinean, 1987; Meacham & Kushner, 1980). However, most of these questionnaires were designed to assess memory ability in general and, therefore, incorporate only a few items pertaining to competence in prospective memory tasks (cf. Hannon et al., 1995). Common examples of questions about prospective memory include, "How often do you forget to keep appointments?," "How often do you forget to take things with you when you go out?," and "How often do you forget something you intended to say?" In general, such prospective memory items tend to get relatively high ratings in terms of frequency of forgetting in daily life. Diary studies typically require subjects to keep a "forgetting journal" for a period of time (e.g., Crovitz& Daniel, 1984; Ellis, 1988; Ellis & Nimmo-Smith, 1993; Terry, 1988). Participants are instructed to carry the journal around with them wherever they go and record all memory failures as soon as they occur. Like the questionnaires, diary studies indicate that a significant proportion of the forgetting that subjects report involve failures in remembering to perform a previously formed intention. In fact, some diary studies indicate that prospective memory failures occur in everyday life as frequently (Crovitz & Daniel, 1984) or more frequently (Terry, 1988) than failures of retrospective memory. Although questionnaire and diary studies provide important information regarding prospective memory abilities (e.g., meta-knowledge), it is necessary to keep in mind that they rely on self-assessment and self-report. Such self-ratings of ability 33 have proved to have high reliability, but quite low validity (see Cohen, 1996; Herrmann, 1984; Taylor, Miller, & Tinklenberg, 1992). Several reasons for this low validity have been identified. Self-assessments may reflect self-image rather than actual performance and may be distorted by such factors as beliefs about the way memory works and response bias. Depression and anxiety also have been found to influence self-assessments (e.g. Rabbitt & Abson, 1990). Differences in opportunity for error may distort results as well. For example, some people may report that their memory for keeping appointments is excellent, but actually make very few appointments (Maylor, 1996b). Another difficulty of self-assessments is that subjects may not be aware of their memory failures. Wilkins and Baddeley (1978) found that subjects were quite good at recalling when they had remembered to carry out a planned action, but were often unaware of when they forgot. This is not surprising because many failures to carry out plans are trivial and can go unnoticed (Meacham, 1982). Another limitation of most questionnaire and diary studies is that they do not reveal how successful prospective remembering is accomplished. Are people successful because they rely on their own memory or because they use external reminders? Success in prospective memory tasks tends to increase with the use of external cues such as writing notes (Harris, 1980; Meacham & Columbo, 1980; but see Maylor, 1990), and, when given a choice, most people prefer to rely on external rather than internal cues when performing prospective memory tasks (Meacham & Singer, 1977). All of these issues indicate that questionnaire and diary studies are unlikely to be very accurate sources of information about prospective memory ability.2 Regardless of their validity, such studies do provide insight into people's perceptions of their own memory abilities, which can have important implications for behaviour (e.g., see Cavanaugh, 1989). 34 (2) Naturalistic field studies. Until recently, naturalistic field studies comprised the bulk of the extant research on prospective memory. Most early researchers investigated memory for future intentions in the real world using common, everyday kinds of tasks. Examples include appointment keeping (e.g., Levy & Claravall, 1977; Levy & Clark, 1980; Martin, 1986), mailing postcards on specific dates (e.g., Meacham & Leiman, 1975/1982; Meacham & Singer, 1977), making telephone calls at precise times (e.g., Maylor, 1990; Moscovitch & Minde, cited by Moscovitch, 1982), and medication adherence (e.g., Park et al., 1999; Park & Kidder, 1996; Wilkins & Baddeley, 1978). As an illustration, in a typical postcard study, participants are given a number of pre-addressed, stamped postcards that they are to mail to the experimenter on appropriate dates. The number of postcards mailed in on time per participant is the dependent measure, and unreturned postcards as well as cards postmarked later than the date specified are taken as evidence of failure in prospective remembering. For a typical appointment-keeping study, the appointment-keeping behaviours of patients at a medical clinic or participants in a psychology experiment are recorded. Prospective memory performance is measured by whether individuals show up for their appointments and approximately how close to the scheduled time they arrive. The major advantage of these studies is that they are practical and closely resemble tasks of everyday prospective memory. They provide valuable information about who is most likely to carry out previously formed plans and intentions in real life situations. Interestingly, most naturalistic field studies have found that older adults are just as good or better at remembering to show up for appointments on schedule, mail postcards on specified dates, and make telephone calls on time than younger adults (see Birt, 1999; Martin, 1986; Maylor, 1990; Moscovitch, 1982; Patton & Meit, 1993; 35 West, 1988). However, as with diary studies, the majority of these studies do not provide any information as to how they remember to perform these tasks at the appropriate times. When memory is tested outside the laboratory, participants are free to adopt any strategy they wish for remembering. Performance advantages may result from the use of external cues or memory aids as reminders. In addition, there is no way to control for or reduce variability in factors such as motivation, compliance, ongoing task demands, distinctiveness of the retrieval cue, etc. in naturalistic field studies. Therefore, the primary limitation of studying memory phenomena in the field is the lack of control over subjects' behaviour. Successful performance on these types of prospective memory tasks does not necessarily mean successful remembering. Similarly, a failure to respond to the prospective retrieval cue does not necessarily mean a failure of prospective memory. Some subjects may remember the intention at appropriate time but choose not to act on it or, for any number of reasons, may be prevented from doing so. (3) Artificial laboratory studies. The majority of the recent research on prospective memory performance has utilized traditional laboratory-type tasks (e.g., Einstein, McDaniel, Richardson, Guynn, & Cunfer, 1995; Einstein, Smith, McDaniel, & Shaw, 1997; Maylor, 1996a; McDaniel & Einstein, 1993). The typical, most frequently employed paradigm for studying prospective remembering in the laboratory is a computer-based dual-task paradigm, first introduced by Einstein and McDaniel (1990). In this paradigm, participants are asked to perform an attention-demanding working memory task, such as memorizing lists of words for short-term recall or answering general knowledge questions, presented on a computer screen. This task serves as the ongoing or "cover" task. For the prospective memory task, participants are 36 instructed to remember to press a computer key whenever a particular word occurs during the ongoing task. The prospective memory cue word appears several times throughout the working memory trials, and the number of times subjects remember to press the key in response to the cue is taken as a measure of prospective memory performance. The greatest advantage of such laboratory studies is the degree of control the experimenter has over the parameters of the experimental situation and subjects' behaviour. All subjects perform the same task under the same conditions, the use of external cues is manipulated and controlled by the experimenter, ongoing task demands are the same for all subjects, and it seems unlikely for there to be important differences in motivation between subjects. Nevertheless, there are several limitations to this experimental paradigm. First, there is often a problem with ceiling and floor effects (e.g., Einstein & McDaniel, 1990; see Kvavilashvili, 1998; McDaniel & Einstein, 1993). Many participants tend to either always remember or always forget to respond to the prospective memory cue, resulting in little variability in the data. In addition, participants tend to respond to the prospective cue immediately or do not respond at all (e.g., Einstein & McDaniel, 1990; Ellis & Milne, 1996). Delayed or late responses are rare. This is most likely due to the fact that there often is no opportunity for the participant to respond late as the task continues to move forward, coupled with the lack of means for measuring such responses. A second limitation is that prospective memory tasks of the type described above do not appear to resemble the kinds of prospective memory tasks that are frequently encountered in daily life. In fact, these dual-performance computer-type prospective tasks often seem to more closely resemble monitoring tasks than what is traditionally thought to be prospective memory tasks. If the subject must keep watch for the appearance of a particular target while 37 simultaneously performing another task, there is no way of knowing how often he or she thinks about the prospective memory task. In order to prevent subjects from thinking about the prospective task, a highly attention-demanding ongoing task (e.g., working memory task) is usually employed. A demanding ongoing task may decrease the likelihood that subjects think about the prospective task, but given that subjects are almost always aware that their prospective memory ability is being tested, it is possible that they modify their behaviour so that they try harder to remember to respond to the cue at the appropriate time. In fact, Kvavilashvili (1998) found that participants who are aware that their prospective memory is being tested commit fewer prospective memory failures than those who are unaware. (4) Naturalistic laboratory studies. Naturalistic laboratory studies refer to those studies that employ a prospective memory task that seems to occur naturally within a laboratory setting (e.g., Dobbs & Rule, 1987; Kidder, Park, Hertzog, & Morrell, 1997; Kvavilashvili, 1987, 1998; Uttl et al., 1999; West, 1988). These type of studies are carried out in such a way that, from the subjects' perspective, the prospective memory task does not appear to be a main focus of the study. In fact, it is often designed to appear quite incidental and completely unrelated to what is being tested. For example, Kvavilashvili (1987) tested subjects by using two laboratory rooms and two experimenters. When subjects were moving from one laboratory to the next they were asked casually if they wouldn't mind asking the second experimenter for the data from another subject, and to bring the data back to the first laboratory when they returned. Similarly, West (1988) had subjects perform a prospective memory task in the context of a memory interview. He extemporarily asked subjects to remind him to make a phone call after he said "the interview is over now." In both of these 38 experiments subjects were scored according to whether or not they performed the prospective memory task. Kidder et al. (1997) had participants complete a number of paper-and-pencil cognitive tests (e.g., vocabulary test) and requested they remember to write the day of the week on the top right-hand corner of every sheet of paper "in order to help the experimenter keep them organized." In this case, prospective memory performance was measured by the number of times the subjects wrote the day of the week on the top of the page and they were scored as successful as long as they wrote it anywhere on the page. The advantage of the naturalistic laboratory study is that the experimenter maintains control over subjects' behaviour while employing naturally occurring tasks that more closely resemble the types of intentions and plans often encountered in everyday life. The experimenter's request is likely to elicit approximately similar levels of motivation within all subjects, especially since they usually are unaware that they are being evaluated on their performance. Given that the prospective task is introduced very naturally and incidentally in the context of ongoing tasks and is not connected with the main experimental instruction, subjects are less likely to be thinking about the intention while they are performing the "main" (i.e., ongoing) experimental task. Single-intention, repeated-intention, and multi-intention designs. Previous studies of prospective memory also differ according to how many measures of prospective memory performance are taken (Ellis, 1996; Ellis, Kvavilashvili, & Milne, 1999; see Kvavilashvili, 1992; Rabbitt, 1996). Single-intention designs include only one measure of prospective memory. The subject is asked to carry out the intention only once (e.g., Dobbs & Rule, 1987; Kvavilashvili, 1987; West, 1988). Repeated-intention designs require that the subjects carry out the same intention several times 39 (e.g., Einstein, Smith, McDaniel, & Shaw, 1997; Einstein & McDaniel, 1990; Ellis et al., 1999; Maylor, 1996a; McDaniel & Einstein, 1993). In multi-intention designs, the subject is asked to remember to carry out two or more different intentions at various times (e.g., Cockburn & Smith, 1988, 1991; Kvavilashvili & Chitashvili, 1991; Somerville, Wellman, & Cultice, 1983; Uttl et al., 1999). Such experimental designs result in different kinds of data. Single-intention designs provide categorical data. Did the subject remember the intention or not? By contrast, repeated- and multi-intention designs allow quantitative measurement. This enables researchers to analyze the prospective memory scores in many different ways. For example, researchers can look at the actual number of remembered intentions, the proportion of remembered intentions, the mean overdue response time, amount of forgetting, contingency scores, etc. Of course, it is important to note that a score computed from only two opportunities for prospective remembering is much less reliable than a score calculated from a greater number of instances of remembering intentions (see Kvavilashvili, 1992). Further, there is a potential problem of dependence between individual measures in repeated-intention designs (as well as many multi-intention designs). That is, excluding the first trial, performance on the prospective memory trials may be influenced by the performance of previous trials. After the first trial, each subsequent performance is likely influenced by prior trials, and as a result, these trials may be confounded with retrospective memory. One way of dealing with this is to compute a separate prospective memory score for performance on the first, uncontaminated trial in addition to computing a composite score. Maylor (e.g., 1990; 1993a; 1998) handles this problem by computing forgetting and recovery scores. Forgetting scores are defined as a failure after a success on consecutive trials, and recovery scores are defined as a success after a failure on the previous trial. Scoring 40 the data in these ways allows for a more in-depth, fine-grained analysis of prospective memory performance than does a single composite score alone. There is some evidence that when choosing between single-intention, repeated-intention, or multi-intention designs, the specific prospective memory task/ experimental paradigm to be employed should be considered. Ellis et al. (1999) investigated the effect of cue presentation frequency on prospective memory task performance using a repeated-intentions design. Interestingly, overall, prospective memory performance was insensitive to variations in event-cue frequency (with number of presentations ranging from 5 to 30). However, different patterns of results were observed with different prospective memory tasks. When the cues were embedded in a general knowledge task, performance improved over cue presentations (i.e., there were practice effects). However, when the cues were presented in the context of a story reading task, performance showed a small decline across presentations. These results suggest that a repeated-intentions design may be more suitable for some prospective memory tasks and for some experimental conditions than others. Issues of Measurement The label prospective memory is often used interchangeably to refer to a type of memory task and a particular task component or requirement, but, as mentioned previously, the distinction between tasks and task components is critical. Researchers have long recognized that it is important to distinguish between two distinct components in prospective memory tasks--the prospective memory component and the retrospective memory component (e.g., Einstein & McDaniel, 1990, 1996; Kvavilashvili, 1987; Maylor, 1996b; Winograd, 1988). Despite this recognition, the 41 majority of researchers have focused on overall prospective memory task performance rather than on the prospective memory task component to measure performance. This is largely due to the fact that overall task performance is relatively easy to measure. Did the subject press the response key or not? Did they show up for their scheduled appointment? Did they make the telephone call at the appropriate time? Although these are clear-cut behaviours that are easy to measure and score, they are not necessarily accurate measures of prospective memory retrieval. All of these behaviours rely on both successful prospective and retrospective memory performance. I could not show up for my appointment on time if I did not remember where or when it was. I could not return a phone call if I did not remember the telephone number or to whom I was to make the call. In these cases, memory for the intention is confounded with memory for the content. It could have been the case that I remembered to make the call but forgot the phone number or simply decided not to bother calling. How should prospective memory retrieval be measured? Is there a way to measure it directly, without relying on what has been termed the "retrospective memory component" of the task? Several suggestions have been made concerning how to more accurately measure prospective memory retrieval. One involves restricting the type of prospective memory task employed, a second involves minimizing the influence of the retrospective memory component, and a third suggests the use of a lenient scoring criterion for successful prospective memory performance. All three suggestions provide useful information about how to measure prospective memory performance more precisely, but it is likely that the combination of the three will yield the most sensitive measure. 42 (1) Restrict the type of task. Ellis (1996) argued that the use of the term prospective memory is too narrow, pointing out that delayed intentions can include things that we should do, things that we need to do, things that we must do, want to do, hope to do, wish we could do, etc. However, although similar, intentions can be differentiated from hopes and wishes. According to James (1892/1961) we all desire to feel, to have, or to do, many different things which at present are not felt, had, or done. If we have the knowledge or belief that attaining what we desire is not possible or very unlikely, we simply wish. If we believe that the fulfillment of the desire is possible, but is not entirely within our control, we hope. However, if we believe that the desire is achievable and within our control, we form an intention to make that desired feeling, having, or doing real. Hoping or wishing to do something in the future is quite different from actually intending to do it. If such a broad approach to the study of prospective memory is taken, then it is clear that it involves much more than the study of memory. Emotion, motivation, meta-knowledge, planning, problem solving, and all the other non-mnestic, "noncognitive factors" discussed previously become much more crucial in determining whether previously formed intentions will actually be carried out. Studying how and why all kinds of intentions are fulfilled is definitely an interesting and important research pursuit, however if the goal is to study memory processes per se, it is probably best to start with a narrow definition of prospective memory (Crowder, 1996; Roediger, 1996). Roediger (1996) argues for limiting the study of prospective memory to "those tasks that a person feels compelled to do, the ones he or she must do" (p. 153). If we study only those intentions that subjects feel they must do, then we can be more certain that a failure to carry out the intention is due to a failure in prospective memory 43 functioning. This has been the reasoning used to guide much of the past research in this area. Nevertheless, although this approach minimizes the chances that an intention is remembered but for some reason not carried out, it by no means guarantees it. In addition, it focuses on overall prospective memory task performance rather than on the performance of the prospective memory component of the task. A narrow definition of prospective memory is definitely a good starting point, but from there the challenge becomes devising a way of measuring more directly whether an intention was remembered, irrespective of whether or not it was actually carried out. (2) Minimize the retrospective component. The approach that Einstein and colleagues (e.g., Einstein & McDaniel, 1990; 1996; Einstein et al., 1995; Einstein et al., 1997; McDaniel & Einstein, 1993) use for increasing the accuracy in measuring the prospective memory component of prospective memory tasks is to minimize the influence of the retrospective component. Essentially, this is accomplished by making the retrospective component of the task as simple as possible and ensuring that all subjects show perfect recall. For example, if subjects were asked to press a key whenever they see the word "book" during the experiment, the experimenter has the subjects repeat this instruction to ensure it was encoded, and at the end of the experiment asks them what it was they were supposed to do to make sure that they did not forget. Most subjects are able to recall what they were supposed to do, and those who do not are excluded from the study. Thus, any failures to respond to the word "book" during the experiment are most likely the result of failures in the prospective memory component of the task. By making the retrospective component of the prospective memory task as easy as possible and by ensuring that subjects do/ did not forget that component, this approach comes closer to a "pure" measure of prospective 44 memory ability. However, a shortcoming of this measurement technique is that it still relies on overall performance on the prospective memory task. Either the task is completed or it is not. This approach alone does not allow for responses in which participants remembered the intention, but did not bother to carry it out or remembered it too late to respond on time. It seems that such response scenarios are possible since the ongoing task is usually designed to be demanding. Participants may not want to interrupt their performance on the ongoing task or they may simply want to wait several seconds to finish what they are doing before they press the response key, especially if they believe the ongoing task is the primary and most important task of the experiment. (3) Lenient scoring criterion. Most approaches to measuring prospective memory performance, like the two mentioned above, use what can be referred to as a strict scoring criterion. Participants are considered successful if they carry out the prospective memory task exactly as instructed. If they do not, they are scored as having forgotten. Dobbs and Rule (1987) were among the first researchers to measure prospective memory performance using a more lenient scoring criterion. They gave subjects a questionnaire to be filled out at home and stressed the importance of putting the time and date in the upper left-hand corner. They then measured prospective memory performance using both a strict and a lenient scoring criterion. The strict criterion required that both the time and the date were written in the correct location, whereas the lenient criterion allowed for either the time or the date to be written in any location. Such a lenient criterion yields a more direct measure of the prospective memory component of the task by providing some measure of remembering to do something and by not requiring that all of the content of the task be remembered. The 45 results of the Dobbs and Rule (1987) study were very different depending on the scoring criterion used. They found that scores calculated using a strict scoring criterion showed only a marginal influence due to age, whereas scores based on a more lenient criterion showed a significant age-related performance decline. In addition, performance scores were higher overall when the scoring criterion was lenient than when it was strict. Therefore, the manner in which prospective memory test performance is scored can have a significant impact on the research findings. Subsequent empirical investigations have incorporated lenient scoring criteria into their task performance measurement schemes. For example, Cockburn and Smith (1994) measured performance on a prospective memory task, which required participants to remember to ask about an appointment, according to five different response options: (1) spontaneously asking about the appointment at the appropriate time; (2) spontaneously asking about something else at the appropriate time; (3) spontaneously asking for something, but not remembering what it was; (4) responding correctly after prompting; and (5) not responding at all or responding incorrectly, even after a prompt. Similarly, Mantyla and Nilsson (1997) scored prospective memory performance according to the following categories of responses: uncued recall (correct response at the appropriate time), cued recall (correct response with a prompt to remember), and content recall (remembering the content of the intention after a reminder). Other researchers (e.g., Kidder et al., 1997; Kvavilashvili, 1998; Maylor, 1993a; Uttl et al., 1999) also have utilized more lenient and detailed scoring systems which attempt to measure the wide variety of possible responses and response errors. Despite its advantages, the use of a lenient scoring criterion is not a perfect measure of prospective memory. Probably the greatest problem with using a lenient score is that it relies on subjects' initiative to indicate that there is something that they 46 should do although they forget what it is that they are supposed to do. That is, it is not clear that subjects who remember that something has to be performed but forget what it is that they have to do will necessarily indicate this to the experimenter In addition, subjects who fail to carry out the task and fail to indicate that they remember something was to be done are often counted as failing the prospective component of the task. It is possible that those subjects simply forgot the task entirely. Despite these problems, the benefit of this method for measuring prospective memory performance is that it enables a more direct and detailed scoring of prospective memory retrieval and decreases the probability of floor effects. By allowing for a wider range of correct responses such as "oh yeah, I think there is something I'm supposed to do now" and delayed or incomplete responses, lenient scoring permits more direct access to, and measurement of, the prospective memory component in prospective memory tasks. Until more sensitive means of measuring prospective memory retrieval are developed, utilizing a detailed/ lenient scoring criteria is a good first step toward attempting to isolate this memory process and gaining further insight into the range of possible responses to prospective memory cues. Definitions and Methodologies: A Summary Implicit in the term "prospective memory" is the assumption that it is a unique form of memory, distinct and dissociable from the various forms of retrospective memory. However, this assumption has yet to be empirically validated. A critical distinction to bear in mind when attempting to define and measure prospective remembering is the difference between memory tasks and memory task components. Prospective memory tasks can be broken down into a number of task components or component processes, only one of which is concerned with prospective memory 47 retrieval. If the mental processes underlying prospective memory retrieval prove to be distinct from those underlying retrospective memory retrieval, it is unlikely that it will be due primarily to the fact that the to-be-remembered content is an intention or that the memory is directed toward the future. A close inspection of the basic task features or component processes of prospective versus retrospective memory tasks reveals that the key difference lies within the roles of the retrieval cues and whether subjects are in a "retrieval" mode when they notice those cues. There appears to be a difference in the availability and salience of retrieval cues. Retrospective memory retrieval is typically explicitly prompted and guided by cues in the environment (e.g., experimenter), whereas prospective retrieval requires unprompted cue recognition. In addition, explicit retrospective memory retrieval is accompanied by an awareness of the connection between the cue and the previous study episode. However, for prospective memory retrieval, it is up to the individual to recognize naturally occurring cues in the environment (mental or physical) as signals of a previously formed intention without being in a mode for recollecting prior experiences. It is this unprompted cue recognition that is the unique feature of prospective memory retrieval because once the cue has been recognized as representing the fact that something should be done, a deliberate search of memory is then initiated to associate the intention with the prior experience of forming it in the first place and recollect what (when and where) actually was intended. Scant research has directly examined the relationship between what has been referred to as prospective and retrospective remembering, and few conclusions can be drawn from that which has been conducted (see Burgess & Shallice, 1997; Dalla Barba, 1993). Some studies have found little or no relationship between measures of prospective and retrospective memory abilities (e.g., Einstein & McDaniel, 1990; 48 Kvavilashvili, 1987; Wilkins & Baddeley, 1978; Uttl et al., 1999), whereas others suggest many similarities between prospective and retrospective memory task performance (e.g., Rendell & Thomson, 1999; Roediger, 1996; Tombaugh, Grandmaison, & Schmidt, 1995). It is apparent that these contradictions are largely to the lack of clear, accurate operational definitions and methodologies for studying prospective memory and differing levels of analysis. In the previous review it was argued that both prospective and retrospective memory tasks can come in many different forms (e.g., short-term, long-term, episodic, habitual, self-generated, other-generated) and task characteristics can vary greatly. Also highlighted was the fact that performance on both prospective and retrospective memory tasks can vary substantially depending on such task characteristics. The terms "prospective" and "retrospective" are umbrella terms for a whole variety of memory tasks. In this dissertation, the focus will be primarily on memory tasks that are of an explicit, episodic nature. Also important to note is that, at present, the term prospective memory refers as much to a specific type of experimental paradigm as it does to a hypothetical mental ability (Burgess & Shallice, 1997; Craik & Kerr, 1996; Dobbs & Reeves, 1996). Recognizing that memory tasks require a whole complex set of mental processes, most generally classified as encoding, "storage", and retrieval, the term prospective memory will be used primarily to refer to the retrieval processes—the act of remembering. In addition, the term prospective remembering will be used to refer to those phenomenological experiences and behaviours involved in remembering previously formed intentions. CHAPTER THREE What Evidence is Needed to Argue for Functional Independence 50 The Relationship Between Prospective and Retrospective Remembering Memory has long been recognized as including distinct types of activities. From at least as early as the time of the ancient Greeks (see Herrmann & Chaffin, 1988), many theorists believed there was a difference between the fleeting and transient nature of memory for information recently encountered and the more enduring and permanent nature of memory for information experienced more often and over longer periods of time. The famous psychologist, William James (1890), distinguished between primary and secondary memory. He maintained that primary memory consisted of the momentary contents of consciousness; that is, thoughts under active consideration. Secondary memory consisted of information that was not active in conscious awareness, but could be brought to consciousness at will. James' distinction between primary and secondary memory corresponds closely to the more modern conceptualizations of short-term and long-term memory (e.g., Atkinson & Shiffrin, 1968; Baddeley, 1986; Waugh & Norman, 1965), which have received an abundance of research attention. Many other distinctions have been made between different activities or tasks of memory, giving support to the notion that memory is not unitary, but instead, involves different processes, functions, or systems that are engaged differentially depending on task properties and demands (see Foster & Jelicic, 1999). Probably the most widely recognized categorization of memory is the episodic, semantic, and procedural trichotomy (Tulving, 1985a). Episodic memory refers to knowledge for personally experienced events including the time and place in which they happened, whereas semantic memory refers to encyclopedic or factual knowledge of the world that is not associated with the autobiographical information of time and place of acquisition 51 (Tulving, 1972; 1983). Procedural memory concerns perceptual, motor, and cognitive skills and habits that are inaccessible to conscious awareness and acquired and demonstrated by doing (Tulving, 1985a). Because episodic and semantic memory both pertain to knowledge of facts, it has been argued that there is no need to distinguish between them (e.g., Cohen & Squire, 1980). As a result, a distinction between declarative and nondeclarative memory was proposed (e.g., Cohen & Squire, 1980; Squire, 1987). Declarative memory refers to factual knowledge (both episodic and semantic) that is directly accessible to conscious awareness, whereas nondeclarative memory is concerned with knowledge that is gained incrementally and not accessible to conscious awareness, including perceptual-motor skills, habits, conditioning, and priming (Squire, 1987; Squire, Knowlton, & Musen, 1993). Memory tasks also can be classified as either explicit or implicit (or direct vs. indirect). Explicit memory tasks require that information from a previous event or experience be recollected in a conscious, intentional manner. By contrast, implicit memory tasks reveal the influence of a prior episode or experience in the absence of any conscious, intentional attempt at recollecting that episode (Graf & Schacter, 1985, 1987). With all of these conceptualizations and classifications of retrospective memory, where does prospective memory fit? Are the same mental processes involved in remembering to do things in the future as are used to remember things that have been done in the past, or are different processes involved in each? Are prospective memory tasks yet another distinct category of memory tasks/ activities or can they be classified under one of the existing categories of retrospective memory? Prospective memory retrieval (as defined here) appears to be most similar to episodic memory retrieval (Wheeler, Stuss, & Tulving, 1997). Memory for an intention is usually accompanied by the awareness that the intention was formed at an earlier time, and therefore can be 52 considered "episodic" (see Graf & Uttl, 1999). However, episodic memory also has been subdivided into a number of distinct memory activities/ classifications, including free recall, cued recall, and recognition. How is unprompted cue recognition related to these types of episodic memory activities? How is it related to other types of memory tasks, such as semantic or implicit tasks? Criteria proposed for identifying distinct memory functions or systems. The basic evidence required for establishing independence between performance on different memory tests has been a matter of intense debate, particularly between systems and process theorists of memory (see Foster & Jelicic, 1999 for a review). The details of the systems versus processes debate is beyond the scope of this thesis (in fact, some argue that there really is no debate; e.g., Tulving, 1999b). However, given that a primary purpose of this dissertation is to determine the nature of the relationship between prospective and retrospective memory retrieval, the different criteria proposed to postulate distinct memory systems/ functions will be discussed briefly. Sherry and Schacter (1987) proposed four main criteria that must be satisfied in order to ascertain a difference between two memory "systems": (1) functional dissociations between performance on the memory tests; (2) stochastic independence between the tests; (3) different neural pathways for the memory systems; and (4) functional incompatibility between the tests. 3 Although evidence for any one of these criteria is usually sufficient to suggest the possibility of a distinct memory system, the confident assertion of an independent system requires that all four be met (even though this is rarely the case). Depending on one's theoretical orientation and preference, the term "system" can in most instances be replaced with process, task, or function throughout the following discussion. After the different criteria for postulating 53 distinct memory systems are outlined, each criterion will be discussed in terms of its applicability to prospective memory tasks. Functional independence (or dissociation) is the most typical form of dissociation and one of the most important criteria for the postulation of independent memory tasks, functions, or systems. A functional dissociation occurs when an independent variable has differential effects on two or more memory tests. The independent variable can be an experimental manipulation, such as level of processing, generation, or modality of presentation, or it can be a subject variable, such as age or brain injury. For example, variations in levels of processing during encoding (deep versus shallow encoding) typically have a substantial effect on explicit memory tests such as recall and recognition, but little or no effect on implicit tests such as word stem completion or perceptual identification (e.g., Jacoby & Dallas, 1981). In addition, whereas explicit memory test performance tends to decline with age, implicit test performance is relatively well preserved (see Graf, 1990). Snodgrass (1989) outlined the variety of functional dissociations that can occur between tests of memory. A single dissociation occurs when one independent variable has different effects on two memory tests, whereas a double dissociation occurs when two independent variables have differential effects on each of two memory tests. Dissociations also can vary according to the level or degree of dissociation observed. According to Snodgrass' (1989) conceptualization, a normal dissociation is one in which an independent variable has an effect on one memory test but no effect on the other. A strong dissociation occurs when an independent variable affects one memory test in one direction and affects the other in the opposite direction. A weak dissociation refers to the situation in which an independent variable affects two 54 different tests in the same direction, but the effect is stronger in one test than in the other. Stochastic independence refers to the relation between performance on two tests, as aggregated across subjects and items. If performance on the two tests is uncorrelated, then the tests are assumed to be independent. Again, the degree to which the tests are dissociated can vary (Snodgrass, 1989). A normal dissociation occurs when there is no correlation between scores on the two tests; performance on one test does not predict performance on the other. A strong dissociation occurs when there is a negative correlation between the tests; successful performance on one test is associated with poor performance on the other. A weak dissociation occurs when there is a positive correlation between the two tests; that is, performance is not dissociated between the two tests. As an example, Tulving, Schacter, and Stark (1982) obtained stochastic independence between word recognition and primed word fragment completion. They found that recognizing a word from a previous study list did not predict priming for the same word on a subsequent fragment completion test. They concluded that this dissociation provided evidence for separate memory systems governing performance on word recognition and implicit word fragment completion tests. Although many investigators have found similar results with different implicit tests (e.g., Jacoby & Witherspoon, 1982), some have had difficulty replicating these results, finding dependence between implicit and explicit tests (e.g., Graf & Mandler, 1984; Greene, 1986). As a result, there has been considerable discussion about the methodological and statistical issues that complicate such analyses (e.g., Hintzman & Hartry, 1990; Howe, Rabinowitz, & Grant, 1993). Thus, at present, using stochastic independence as evidence for distinct memory systems or functions, although 55 necessary and quite informative, is not sufficient for making definitive conclusions about the nature of relationships between performance on memory tests. One difficulty with using functional and stochastic dissociations as evidence for independent memory systems is that dissociations have been found between tests that are supposedly governed by the same memory system. For example, Blaxton (1989) demonstrated a functional dissociation between performance on two episodic memory tests (semantic and graphemic cued recall) and between two semantic memory tests (word fragment completion and general-knowledge questions). In addition, Hayman and Tulving (1989) found a stochastic dissociation between two successive tests of implicit word fragment completion that differed only in the fragments presented (e.g., assassin, a a in vs. _ss_s n). Performance on the primed fragment completion tests was correlated only if the same fragments were presented on both tests. If different fragments were used between the tests, strong dissociations resulted. In contrast, stochastic dependence was observed between successive tests of fragment cued recall, regardless of whether the fragments were the same or different between the tests. How have within-systems dissociations been dealt with? Some theorists have handled such inconsistent findings by postulating new memory systems for the dissociated tasks (see Roediger, 1990). For example, Tulving and Schacter (1990) introduced a new memory system called the perceptual representation system (PRS) to account for dissociations between tasks previously thought to be guided by semantic memory. Performance on tasks such as word fragment completion, word stem completion, and lexical decision is now argued to be dependent on this perceptual, presemantic system. However, an important point to keep in mind with regard to such dissociations is that human information processing is extremely complex and as a 56 result one can never get a "pure" measure of any memory process or system. This difficulty in isolating mental processes has been referred to as the "process-pure" problem (Jacoby, 1991; Toth & Hunt, 1999). Most memory tasks involve more than one activity of memory. For example, although explicit memory tasks are thought to require primarily episodic memory functioning, they likely reflect contributions from semantic memory functioning as well. In addition, it is difficult to determine the degree to which episodic memory contaminates performance on an implicit memory test. Thus, how can one be sure that dissociations at the level of tasks reflect the working of dissociable, underlying memory systems/ functions, especially when performance on the tests is usually the result of contributions from more than one memory system or function? The answer appears to lie in carefully designing memory tasks and employing methodologies that reduce confounding variability and isolate the underlying memory processes as much as possible, and in finding evidence for converging dissociations across different versions of memory tasks and across different populations (see Foster & Jelicic, 1999). A third criterion argued as necessary for postulating separate memory systems is the reliance on different neural pathways. The neural processes responsible for the cognitive operations required for performing the memory tasks should differ. Although no one today would argue against the fact that brain activity produces memory activity and although technological advances have vastly improved the study of brain activity, fulfilling this criterion has been somewhat of a challenge. One problem is associated with the lack of clear definitions. Terms such as memory systems, brain systems, brain processes, neural structures, neural pathways, neural mechanisms, etc. are poorly defined in both psychology and neurology (Weldon, 1999). A second problem with the criterion of distinct neural pathways concerns the 57 issue of treating psychological functions, such as memory, as isomorphic with brain structures. According to Toth and Hunt (1999), this problem is primarily concerned with how one defines memory in the first place. Is memory simply neuronal activity? Or, is memory a dynamic psychological phenomenon or behaviour reflecting the interaction of a particular individual with a specific prior history in a specific task environment? The way in which memory is defined will determine the level of analysis. A third problem with the different neural pathways criterion is the issue of what makes one neural pathway different from another (Roediger, 1990; Roediger, Buckner, & McDermott, 1999). How much or how little similarity is acceptable in order for two neural pathways to be declared different? One would expect that the neural systems and processes would be different for memory tests that are quite different in their performance requirements, for example, primed homophone spelling and word recognition. However, it is possible to equate memory tests on their task features and requirements such that the only feature that differs is the instructions (e.g., Graf & Mandler, 1984; Schacter, Bowers, & Booker, 1989). For example, an implicit word stem completion test and an explicit cued recall test that use the same target word (triumph) and the same test cue (tri ). In producing the same word (triumph) for both tests, it seems likely that many of the neural processes and systems would be the same. To what extent do the neural pathways have to be different in order to qualify as different systems? This question remains to be answered. Nevertheless, at present, neuropsychological dissociations are typically considered as representing strong evidence for the possibility that performance on two memory tasks may be governed by independent functions or activities of memory (e.g., Burgess & Shallice, 1997; Cockburn, 1995; McDaniel eta l . , 1999; Schacter & Tulving, 1994; West, Herndon, & Ross-Munroe, 2000). 58 Functional incompatibility is the fourth major criterion believed to be important for postulating distinct memory systems. The main idea behind the functional incompatibility criterion is that memory systems are specialized to such an extent that the functional problems each system handles cannot be handled by another system. One way that memory systems are purported to be specialized is in terms of the kinds of contents or forms of information they handle and store (e.g., Schacter & Tulving, 1994). For example, episodic memories refer to autobiographical event knowledge, whereas semantic memories refer to facts and general knowledge. According to functional incompatibility, a system that is specialized to handle one specific type of content would be unable to process a different kind of content. Using an evolutionary perspective, Sherry and Schacter (1987) argued that different memory abilities evolved as adaptive specializations in response to demands in the environment that could not be handled by existing memory abilities. Although most memory investigators agree that the criterion of functional incompatibility appears plausible, especially with regard to how different activities of memory may have evolved, in reality it has rarely been applied to human memory systems. The main challenge in fulfilling this criterion is that it is difficult to define exactly what constitutes functional incompatibility and the conditions under which a new memory system would be required and develop (Roediger et al., 1999; Weldon, 1999). For example, why would deliberately remembering an event from the past (episodic memory retrieval) be functionally incompatible with the experience of that event influencing behaviour in the absence of any deliberate attempt at remembering it (implicit memory retrieval)? Similarly, why would remembering experiences that happened in the past be incompatible with remembering intentions for the future? Intuitively, it seems that memory retrieval would work more effectively and efficiently if the "systems" worked together in an integrated 59 fashion rather than being independent and functionally incompatible (Weldon, 1999). Also, even if different memory tasks have different performance requirements, why could the principles of memory not be the same for those tasks? One such principle of memory that seems to cut across the gamut of memory tasks is the principle of transfer appropriate processing. As this principle is important for understanding memory performance, it warrants some discussion. Transfer-appropriate Processing Regardless of how many distinct memory systems or processes there really are or whether activities of memory should be considered as systems, processes, or functions, there appears to be one general principle according to which performance on a wide variety of memory tasks complies-the principle of transfer-appropriate processing. This framework, rooted in Kolers' procedural view of memory (e.g., Kolers, 1975, 1976), Tulving's encoding specificity hypothesis (Tulving & Thomson, 1973), and the work of Morris and colleagues (Bransford, Franks, Morris, & Stein, 1979; Morris, Bransford, & Franks, 1977), does not focus as much on memory tasks themselves as it does on the match in mental operations performed during encoding and retrieval (study and test). According to the logic of transfer-appropriate processing, memory test performance will be facilitated to the degree that the same, or similar, types of mental operations are used as those engaged during previous study. In other words, memory performance is determined by the extent of overlap between study and test processing. The human memory literature is replete with validations of transfer-appropriate processing (see Roediger & Srinivas, 1993; Roediger, Weldon, & Challis, 1989). For example, memory retrieval has often shown context dependence (e.g., Davies & Thomson, 1988; Godden & Baddeley, 1975), state dependence (e.g., Eich, 1980, 60 1989), and even mood dependence (e.g., Bower, 1981; Eich & Metcalfe, 1989). In addition, priming is generally larger for words that are studied and tested in the same as opposed to a different sensory modality (e.g., visual-visual versus auditory-visual) (e.g., Schacter & Graf, 1989), when study and test items are presented in the same rather than different symbolic forms (e.g., words versus pictures) (e.g., Weldon & Roediger, 1987), or in the same versus different languages (e.g., Watkins & Peynircioglu, 1983). Transfer appropriate processing is such a pervasive phenomenon there is little reason to believe it would' not apply to prospective memory performance as wel l -a finding that would suggest similarity between prospective and retrospective memory. McDaniel, Robinson-Riegler, and Einstein (1998) examined prospective memory test performance as a function of a number of manipulations in the degree of overlap between encoding and retrieval conditions. They found that changing the semantic context (using homographs) between study and test significantly attenuated prospective memory performance. They also found that varying the symbolic form (words versus pictures) in which the prospective memory targets were presented influenced prospective remembering. Although they found a picture superiority effect, there was a clear performance advantage overall when the symbolic form of the target was the same across study and test. Meier and Graf (2000) manipulated the level of processing (semantic or perceptual) required during the ongoing activity within a prospective memory task. They also manipulated the level of processing (semantic or perceptual) required for prospective memory retrieval (i.e., recognizing the prospective memory target cue as a sign of an intention). In accordance with transfer appropriate processing, memory performance was higher when the mental processing necessary for carrying out the ongoing task and for prospective retrieval were the same than 61 when they were different (see also Einstein et al., 1995; McGann, Ellis, & Milne, 2000; cf. Darby & Maylor, 1998). These findings indicate that prospective memory performance appears to be governed by the same general principle, transfer-appropriate processing, as retrospective memory performance and offer support to the notion that prospective and retrospective memory retrieval may be more similar than they are different. It is important to note that although transfer appropriate processing serves as a good guiding principle for memory research, it does not by itself provide an explanation for performance dissociations between different memory tasks (or lack thereof). What it does provide is a useful framework for thinking about such dissociations and memory task performance in general. It is an assertion about human memory that serves as an organizational principle for theorizing about specific performance dissociations but does not make any special claims about explaining their existence. However, by following the logic of this framework and using it as a starting point in the investigation of memory phenomena, it may be possible to gain valuable insights into the relationship between prospective and retrospective task performance. It is apparent from the review above that the most basic evidence required to even postulate that prospective and retrospective memory retrieval are functionally distinct are demonstrations of independence (functional and stochastic dissociations). Other criteria have been proposed as necessary for identifying unique memory "systems", such as the criteria of separate neural pathways and functional incompatibility, but their assumptions and practical utility are still a matter of debate. At present, there is no "golden rule" or step-by-step checklist for determining whether a specific piece of evidence is sufficient for making solid claims about the independence of processes responsible for performance on two memory tasks. Nevertheless, 62 demonstrations of converging dissociations utilizing different forms of tasks and across different populations appear to offer the most convincing evidence. To date, very little research has tested directly the relationship between prospective and retrospective memory. However, a number of studies have included measures of each and compared performance. Before the results of these studies are summarized, an overview of some of the models or frameworks of prospective memory task performance is provided, along with the specific predictions about the relationship between prospective and retrospective memory tasks that follow from each. This discussion will provide a framework for thinking about and interpreting the pattern of findings within the existing literature. CHAPTER FOUR Theoretical Accounts of Prospective Memory Retrieval 64 The "Recognition-Recall" Model Of Prospective Memory Also referred to as the noticing + search model (Einstein & McDaniel, 1996), the recognition-recall model of prospective memory is derived from familiarity-retrieval theories of recognition memory (e.g., Jacoby, 1984; Mandler, 1980). This model posits that the prospective component (recognizing a target event as a cue for a previously formed intention) of prospective memory tasks involves recognition memory processes and the retrospective component (retrieving the content of that intention) requires recall processes (Einstein & McDaniel, 1996; McDaniel, 1995; Robinson-Riegler, 1994). Successful prospective memory performance is proposed to occur when the retrieval cue or context spontaneously or involuntarily elicits feelings of familiarity or perceptual fluency that lead to the recognition of the cue. This cue recognition then prompts a purposeful search of memory in an attempt to recollect exactly what the cue signifies. The basic assumption is that prospective remembering reflects the operation of two distinct sequential processes, the first of which is relatively automatic and the second that is more controlled (Einstein & McDaniel, 1996). Whether or not an intended action will actually be carried out is determined not only by how successful the directed search of memory is, but also by whether this search is initiated in the first place. Cue recognition is dependent on the degree to which the cue and retrieval context evoke a sense of familiarity, a feeling of knowing, perceptual fluency, and noticing—all phenomena that have been proposed to occur automatically (e.g., Jacoby, 1984, 1991; Jacoby & Whitehouse, 1989). According to this model, the unique feature of prospective memory tasks (unprompted cue recognition) is a form of recognition memory. Therefore, performance on prospective memory tests should more closely resemble performance 65 on other recognition memory tests than tests of recall. In fact, research has shown that, similar to performance on recognition memory tests (Mandler, 198,0), prospective memory task performance is higher when the retrieval cues are unfamiliar and distinctive than when they are less familiar and less distinctive within the retrieval context (e.g., McDaniel & Einstein, 1993). However, the type of recognition that occurs in prospective memory tasks differs from the recognition in retrospective memory tasks in an important way. Episodic retrospective recognition tasks explicitly prompt the rememberer to initiate a memory search or judgment, whereas prospective tasks involve no such prompting. Therefore, one might expect the recognition process in prospective memory tests to differ to some extent from the recognition processes in explicit recognition tests. The former appears to involve involuntary remembering, whereas the latter seems to be more voluntary and controlled. The "Automatic Associative Activation" Model Of Prospective Memory Rooted in connectionist and network models of memory (e.g., Anderson, 1983), the automatic associative activation model is based on the assumption that during the encoding phase of prospective memory tasks, the subject forms an association between the target cue and the to-be-performed action (see Einstein & McDaniel, 1996; Goschke & Kuhl, 1993, 1996; McDaniel & Einstein, 1993; McDaniel et al., 1998). This association results in a heightened activation of the cue-target pairing in memory. During the retention interval, the subject begins to perform other activities and the level of activation of the cue-action association decreases to levels below conscious awareness. The activation level continues to decrease over time, and as a result, so does the probability that the prospective memory will be reactivated and enter conscious awareness at a later time. However, to the extent that the cue-action 66 association is rehearsed, its level of activation is increased as well as the likelihood that an encounter with the target cue will raise the level of activation above threshold and result in successful prospective remembering. Thus, whether the target cue will be recognized as a sign of a previously formed intention depends on the activation level of the cue-action association at the time the target retrieval cue occurs. Level of activation depends on the strength of the particular pathway between the target cue and the action as well as how many other pathways are associated with the target. Goschke and Kuhl (1993) have argued that, compared to other types of to-be-remembered events, to-be-performed actions (i.e., intentions) are held in a state of extra-activation in anticipation of retrieval. This heightened activation further increases the likelihood that the intention will be activated in response to the target cue. According to this model, prospective remembering occurs automatically. When the activation level is high enough, an encounter with the target cue automatically triggers the activation of its associated action. If the activation level is not sufficient, the retrieval cue will be ineffective and the reflexive retrieval process will not be stimulated. In contrast to the recognition-recall model of prospective memory, which maintains that prospective remembering depends on both automatic and controlled retrieval processes, the automatic associative activation model contends that only automatic retrieval processes are involved. No directed, purposeful search of memory is required. Therefore, if this model is correct, one would expect performance on prospective memory tasks to be more closely related to performance on other memory tasks in which retrieval occurs automatically. Some memory researchers have argued that the fluent performance observed in implicit memory tests is the result of automatic processing (e.g., Cermak, 1993; Jacoby, 1991; Jennings & Jacoby, 1993). To the extent that performance on implicit 67 tests of memory reflects automatic, fluent, unintentional processing and if the automatic associative activation model is correct, one would expect implicit and prospective memory tasks to be highly related. Prospective and implicit tests also appear to be quite similar in that when retrieval cues are presented, the subject is not informed or explicitly made aware that the cues are associated with a prior experience (i.e., encoding). Thus, the retrieval context does not direct one to think back to the encoding episode to try to remember the target items. However, the major difference between the two tests is that in the case of implicit tests of memory subjects are deliberately alerted to the target retrieval cue, whereas in the case of prospective tests it is up to the subjects to notice the cue on their own. The work of Gollwitzer and colleagues (e.g., Gollwitzer, 1999; Gollwitzer & Bayer, 1999; Gollwitzer & Brandstaetter, 1997) on automatic goal activation and implementation intentions may lend some insight into how some goals, although not necessarily well-practiced or rehearsed, appear to be executed automatically. According to Gollwitzer, individuals can delegate the initiation of goal-directed behaviour to environmental stimuli by forming implementation intentions (if X occurs, I will do Y). He has found that forming such specific implementation intentions facilitates attention to, memory for, and recognition of the critical retrieval cue/context X . Consistent with the automatic associative activation model of prospective memory, he also has observed that when the retrieval cue and context are present, they may be processed automatically and the behaviour of carrying out the previously formed intention is often immediate and does not necessarily require conscious intent (and is sometimes triggered outside the individual's awareness). The circumstances under which strong associations between ideas/ behaviours can be created in memory in the absence of rehearsal and how retrieval cues and context come to be processed 68 automatically need to be explored further within the area of prospective memory retrieval. A Functional Account Of Prospective Memory Building on previous functional accounts of memory (e.g., Bartlett, 1932; Kolers, 1975; Neisser, 1967), Craik (1983, 1986) asserted that memory is a psychological activity or function determined by the relation between internal mental processes and the external retrieval environment. Prior experience becomes manifest as a result of the interaction between the cognitive capabilities and current goals of the rememberer and the specific cues and task demands of the retrieval context. As discussed above, the transfer appropriate processing principle of memory maintains that remembering will be successful to the extent that cognitive activities and environmental context occurring at retrieval are similar to those that occurred at encoding (Kolers, 1975, 1976; Morris et al., 1977; Tulving & Thompson, 1973). However, many retrieval contexts do not bear much similarity to the rememberer's original cognitive processing or environmental circumstances and in these cases it is up to the rememberer to rely on his or her own mental prowess to reconstruct the memory. To account for how memory retrieval is carried out under varying degrees of overlap between encoding and retrieval, Craik (1986) distinguished between mental processing that is guided and supported by the environment versus processing that is self-initiated, effortful, and controlled. He argued that when environmental support is available, for example in the form of retrieval cues, memory retrieval will be guided primarily by the cues. However, when little or no environmental support is present, remembering is heavily dependent upon self-initiated processing (e.g., preexisting knowledge). 69 Craik (1986) proposed that memory tasks can be placed on a continuum with environmental support at one end and self-initiated processes at the other end. Using this continuum, he presented a hierarchy of memory tasks with priming at the bottom, involving the most environmental support, followed by relearning, recognition, cued recall, free recall, and finally, prospective remembering at the top, requiring the greatest degree of self-initiated processing. Prospective memory tasks do not afford much reliance on external retrieval cues. The absence of any direct prompts to remember means that it is up to the individual to recollect the previously formed intention at the right time and/or in the appropriate context. In addition, because prospective memory performance requires the interruption of ongoing tasks and often the modification of habitual routines, it is unlikely that many relevant cues will be present in the retrieval context (Morris, 1992). In other words, prospective memory tasks are probably much less likely to show any significant overlap in mental operations and environmental cues between encoding and retrieval. However, this does not mean that prospective tasks do not vary according to the degree of similarity between encoding and retrieval contexts. For example, event-based prospective memory tasks have an external event that acts as a cue and has the potential to trigger and guide remembering, whereas most time-based tasks do not (at least not as salient of an event). Therefore, it is assumed that time-based tasks depend on self-initiated/ controlled processing to a greater extent than event-based tasks (e.g., Einstein & McDaniel, 1990). Thus, contrary to the automatic associative activation and recognition-retrieval models of prospective memory, the functional account maintains that the cue recognition that is required in prospective memory tasks does not occur automatically, but demands a great deal of effortful, controlled, self-initiated processing. According to 70 the hierarchy of memory tasks proposed by Craik (1986), the mental processes involved in prospective remembering should overlap substantially with those needed for tests of free recall because they are both considered to demand a high level of self-initiated processing to carry out. However, prospective remembering would be expected to share fewer common processing resources with tests of recognition and implicit memory because performance on these tests is much more environmentally driven. To summarize, if the recognition-recall model of prospective memory is correct, then prospective memory performance should be more related to performance on explicit recognition tests than tests of recall or tests of implicit memory. If the automatic associative activation model is correct, prospective memory task performance should show more parallels with implicit memory tests than either explicit recognition or recall. However, if the functional account of prospective remembering is accurate, prospective memory performance should show more similarities with free recall tests than recognition tests, and even fewer similarities with implicit tests of memory. As a first step in determining the nature of the relationship between prospective and retrospective memory retrieval, the results of the existing research studies including both prospective and retrospective memory tasks were summarized quantitatively. The results of this summary are presented in the next chapter. 71 CHAPTER FIVE Prospective and Retrospective Memory Task Performance Compared 72 Evidence For (In)Dependence From Demonstrations Of Functional And Stochastic Dissociations As mentioned, little research has attempted to directly address the question of whether the same mental processes are utilized in remembering to do things in the future as are employed to remember things that have been done in the past. Nevertheless, a number of research studies have included measures of both prospective and retrospective remembering, which provides an opportunity to examine some of the similarities and differences between the two categories of memory tasks. Table 1 provides a summary of the existing research findings, and a more detailed summary of the research is presented in Appendix A. A quantitative review of 44 different empirical studies yielding 148 prospective-retrospective task performance comparisons revealed that there is little convincing evidence to support the argument that prospective memory is a functionally distinct activity or system of memory or the argument that the same processes govern prospective and retrospective remembering. The summary of the findings relating to functional dissociations shows that very few double dissociations have been observed between prospective and retrospective tests. Further, although there does seem to be a substantial number of single dissociations in the research literature, there are as many observations of prospective and retrospective tests showing similar patterns of effects. This is true regardless of whether comparisons are made with tests of free recall, recognition, or short-term memory. The evidence from tests of stochastic dissociation appear somewhat more promising, with the majority of observations (71.7%) indicating little or no correlation between prospective and retrospective memory performance. Again, this was the case for free recall, recognition, and short-term memory tasks. However, because of the 73 fact that the prospective and retrospective memory tests compared almost always had very different test features and properties, these findings should be interpreted with caution. A closer examination of the patterns of dissociations between prospective and retrospective memory tests reveals surprisingly few consistent findings (see Appendix A). For example, variables such as task load, cue familiarity, cue importance, and retention interval resulted in as many similar effects on the two types of tests as differential effects (see also Roediger, 1996). However, the three studies investigating the influence of memory aids (e.g., external cues, reminders) on memory performance all produced dissociations between prospective and retrospective task performance, with external aids facilitating prospective performance but having no influence on retrospective performance. Further, studies examining age differences in memory performance were far more likely to show an advantage for young adults on both prospective and retrospective memory tasks than any differential effects on the tasks. These findings suggest some potential similarities and differences between the two tasks, but by themselves do not indicate any solid trends in the research literature. The data from the correlational approaches, although they slightly favor independence, also are inconsistent. Proportionately, retrospective tests of short-term and recognition memory were more likely to be uncorrelated with tests of prospective memory than were tests of free recall. Specifically, 10 of the 11 (91 %) correlations computed between performance on short-term retrospective and prospective memory tests from the studies reviewed were not statistically significant; 8 of the 10 (80%) correlations computed between recognition memory test performance and prospective memory test performance were not significant; and 20 of the 29 (69%) correlations between retrospective free recall test performance and prospective memory 74 performance calculated did not reach statistical significance. Does this mean that tests of free recall are more similar to tests of prospective memory than short-term memory or recognition tests? Does it mean that the close relationship between prospective and recognition memory tests predicted by the recall-recognition hypothesis of prospective memory is incorrect?-not necessarily. It is important to keep in mind the possibility that the absence of correlations between two memory tasks may be more reflective of differences in task features and parameters related to the experimental investigation of the tasks than any real differences in the processing requirements (see Toth & Hunt, 1999). Indeed, many of the prospective and retrospective memory tasks compared were very different (e.g., a medication adherence task and free recall of a list of words; remembering to press a button every three minutes and recalling the exact task instructions). Such differences reduce the chances that two tests will show a significant positive correlation, even two tests purported to measure the same type of memory. Thus, at this point, little can be concluded about the relationship between performance on prospective memory tasks and retrospective memory tasks such as free recall, recognition, and short-term memory. There are as many inconsistencies across similar types of tasks (e.g., free recall of a word list, free recall of the prospective cues) as there are across different tasks (e.g., free recall, recognition). Out of the 44 studies reviewed, only two included both prospective and implicit tests of memory. In one study (McDaniel & Einstein, 1993), word fragment completion was moderately correlated with an event-based prospective memory computer task (r = .32) and with a more naturalistic lab task-remembering to write the day of the week on top of several response sheets (r = .31). In the other study (Otani et al., 1997), word stem completion was not significantly correlated with an event-based prospective memory computer task (r = .14). However, in the same study, manipulations in task 75 load did not affect performance on either test; that is, there was no functional dissociation. Therefore, out of four comparisons, three provide evidence to suggest that implicit and prospective memory tests may share more similarities than differences. However, research has demonstrated that unlike most implicit tests of memory, prospective memory tests are subject to generation effects (McDaniel, 1995), can show levels of processing effects (Einstein & McDaniel, 1996), produce picture superiority effects, are influenced by semantic context changes, tend to show performance decrements when attention is divided, and are sensitive to environmental context reinstatement (see McDaniel et al., 1998), much like tests of explicit recall and recognition. Nevertheless, as the relationship between implicit and prospective memory performance has yet to be examined directly, no firm conclusions can be made about the degree to which they are associated. Evidence Concerning Distinct Neural Pathways If it is the case that prospective and retrospective memory retrieval are governed by separate, dissociable memory systems, they should rely on different neural pathways or mechanisms. Although anecdotal evidence and clinical observation suggest that problems in prospective remembering are often reported in conjunction with problems of retrospective memory, research testing prospective memory performance in brain damaged patients is sparse. Only eight of the 44 studies including both prospective and retrospective measures of memory reviewed from the literature, tested patient populations (see Appendix A). Within these eight studies, one double dissociation, 13 normal single functional dissociations, and two stochastic dissociations were observed. There were 12 instances in which dissociations between prospective and retrospective tasks were not found. Again, dissociations with 76 prospective memory tests were equally likely to occur (or not occur) regardless of whether the retrospective memory test was recall, recognition, or short-term memory. A number of researchers have postulated that prospective memory task performance relies heavily on frontal lobe functioning (e.g., Bisiacchi, 1996; Burgess & Shallice, 1997; Cockburn, 1995, 1996; Glisky, 1996; Kerns, 2000; Okuda et al., 1998; Shallice & Burgess, 1991; Shimamura, Janowsky, & Squire, 1991). Such postulations are based on the observation that prospective memory tasks consist of component processes that are assumed to be supported by frontal structures. For example, as discussed previously, it has been argued that prospective remembering relies largely on self-initiated retrieval because there are no explicit prompts directing a search of memory. Therefore, it is up to the individual to monitor his or her environment for the occurrence of a cue that serves as a sign of the previously formed intention. Such initiating and monitoring activities are assumed to be associated with frontal functioning (e.g., Burgess & Shallice, 1996; Shallice & Burgess, 1991; Stuss & Benson, 1984). Okuda et al. (1998) conducted a PET study of prospective memory task performance that provides some empirical support for this assumption. Although they did not assess prospective memory retrieval in and of itself, they did find evidence that the activity of holding or maintaining an intention in mind while performing other activities is mediated by frontal lobe structures. Also mentioned earlier is the fact that once the prospective memory cue has been recognized, a purposeful, directed search of retrospective memory is usually triggered in an attempt to think back to the time in which the intention was formed and retrieve its contents. Such controlled, voluntary, strategic processing is also believed to be mediated by the prefrontal cortex (e.g., Shimamura, et al., 1991). Furthermore, cognitive functions such as goal setting, planning, dividing attention, interrupting ongoing activities, inhibiting inappropriate 77 actions, organizing behavioural sequences, and evaluating performance are all presumed to be supported by frontal structures (e.g., Baddeley, 1986; Burgess & Shallice, 1996, 1997; Luria, 1966; 1973; Shimamura, 1994; West, Herndon, & Ross-Munroe, 2000). The ability to mentally project oneself into the future in order to form intentions in the first place has been attributed to the frontal lobes as well (Tulving, 1985b, 1993; Wheeler, Stuss, & Tulving, 1997). Damage to the prefrontal cortex usually does not result in a loss of specific skills, information, reasoning, or problem solving abilities, but instead impairs the programming and organization of behaviour (Baddeley, 1986). Memory disorders associated with lesions to the frontal lobes do not typically involve an impairment in memory functioning per se. Performance on tests of free recall, cued recall, or recognition, considered to be governed by the hippocampal system (e.g., Squire, 1992), is usually intact with frontal lesions (Shimamura et al., 1991), although there have been some reports of impaired free recall abilities (e.g., Jetter, Poser, Freeman, & Markowitsch, 1986). Thus, it would appear that prospective and retrospective remembering rely on distinct neurological mechanisms, with prospective memory retrieval supported primarily by the frontal lobes and retrospective memory retrieval dependent on the hippocampal system (i.e., medial-temporal lobe and hippocampus) (see Moscovitch, 1994; Squire, 1992). However, this distinction may not be so clear-cut. First, prospective memory tasks rely on retrospective memory retrieval, at the very least, to remember the content of the intention (i.e., the intended action). Second, most patients with damage to the prefrontal cortex are unable to spontaneously use context to facilitate memory retrieval. Therefore, to the extent that the cue information or retrieval context is poorly specified in any memory task, whether it be prospective or retrospective, there should be a greater demand for self-initiated retrieval (see Craik, 78 1986). Hence, prospective memory tasks can vary in the extent to which they involve the frontal lobes, and, although most retrospective memory tasks do not require substantial frontal input, they can vary according to degree of frontal lobe involvement as well. Given these speculations about the neuropsychological underpinnings of prospective memory, what kind of relationship would be expected between prospective and retrospective memory task performance? Glisky (1996) outlined several predictions stemming from the frontal lobe hypothesis of prospective memory, including: (1) most standard tests of retrospective memory will not correlate with tests of prospective memory, especially when the retrospective component in the prospective task is minimal; (2) when the retrospective component is more substantial and demanding, then the two tasks will be more likely to correlate; and (3) prospective and retrospective tasks that both rely significantly on frontal lobe functioning should correlate. At present, there is little evidence to support the first prediction. Most experimental tests of prospective memory attempt to minimize the influence of the retrospective component in prospective memory tasks by making it as simple as possible, but, as can be seen from Appendix A, these types of prospective tasks do not appear to be less likely to correlate with retrospective tasks. With regard to the second prediction, research has shown that increasing the retrospective memory load tends to decrease prospective memory performance (e.g., Otani et al., 1997). However, the results have been mixed regarding the extent to which such tasks are correlated with retrospective memory tasks. It is often the case that the retrospective task being compared is very simple, such as recalling or recognizing the short list of prospective memory target cues used in an experiment. Nevertheless, there does seem to be some, albeit slight, evidence to support the third prediction. Given that some tests of 79 retrospective memory, such as free recall, are assumed to rely on self-initiated processing, and hence frontal lobe functioning, to a greater extent than others (e.g., recognition) (see Craik, 1986), one would expect performance on tests of free recall to be more closely related to prospective memory tests than tests of recognition. A summary of the tests of stochastic dissociation from Appendix A indicates that, proportionately, free recall tests were slightly less likely to result in dissociations than recognition tests. Specifically, 69% of the prospective-free recall test correlations coefficients examined revealed evidence for dissociations, whereas 80% of the prospective-recognition test correlations reviewed showed dissociations. Burgess and Shallice (1997) argue that prospective memory task performance necessarily relies on retrospective remembering and involves a special application of the processes used in episodic retrospective remembering. They propose that prospective and retrospective memory tasks should not be doubly dissociable; only a single dissociation between prospective and retrospective tasks should ever be observed. In other words, individuals showing significant impairments in retrospective memory abilities should always perform poorly on prospective memory tasks, but individuals with problems performing prospective tasks will not always have retrospective memory difficulties. A study by Shallice and Burgess (1991) provides partial support for this assertion. They tested three patients with evidence of focal frontal lobe damage on several open-ended planning, multiple subgoal tasks (e.g., Multiple Errands Test, Six Element Test). Although all three patients showed little or no impairment on tests of retrospective memory, they were significantly impaired on the planning tests compared to normal controls. To the extent that these planning, multiple goal tests tap into components of prospective memory tasks, this study provides evidence that it is possible to have poor prospective memory abilities and 80 intact retrospective memory abilities (i.e., a single normal dissociation). A second study, described in Shallice and Burgess (1997), included 30 brain-damaged patients with varying pathologies and examined the relationship between retrospective memory and planning abilities. A significant negative correlation was found between plan-following behaviour (scored inversely in terms of number of errors) and retrospective memory scores in the brain-injured group, indicating that deficits in retrospective memory functioning are associated with deficits in planning/prospective memory task behaviours. Interestingly, no relationship was found between performance on the two tasks for normal controls. Burgess and Shallice (1997) argued that relationships between prospective and retrospective remembering are probably most likely to be observed when studying individuals with a marked impairment in retrospective memory functioning. Cockburn (1996) found similar results with a group of 18 brain-injured patients showing a wide variety of predominant location of lesion. Patients who performed poorly on retrospective memory tests were more likely to fail an event-based prospective memory task than patients who performed satisfactorily on retrospective tests of memory. In order to test whether prospective memory task performance is more dependent on the frontal lobes or the hippocampal system, McDaniel et al. (1999) divided older adults into four groups based on scores on tests that measure frontal functioning and medial temporal lobe functioning. The four groups represented the factorial combination of high and low hippocampal functioning crossed with high and low frontal functioning. Participants were given an event-based prospective memory test and tests of recall and recognition. The pattern of results that emerged with respect to the recognition test resembled a normal double dissociation. Those participants with low scores on frontal functioning showed diminished prospective 81 memory task performance compared to those with high frontal scores, whereas the low and high functioning hippocampal groups did not differ significantly. By contrast, high hippocampals performed significantly better than low hippocampals on the recognition test, but the high and low frontal groups did not differ in their performance. Although these findings suggest that prospective and retrospective memory tasks may be doubly dissociable and that prospective task requirements seem to have little overlap with the requirements of tests of recognition, such conclusions may be premature. First, the number of participants included in the comparisons involving the recognition and free recall were very small-only four to six in each cell. Also, although not statistically significant, there was a 17% advantage in prospective memory performance for high hippocampal participants relative to low hippocampal participants. Thus, as McDaniel et al. (1999) noted, the failure to find a statistically significant effect may be more of a methodological problem than any real absence of a difference between the groups. An important point that needs to be reiterated here concerns the distinction between memory tasks and memory task components. Most of the theorizing, predicting, and even the research conducted attempting to determine the degree of frontal lobe involvement in prospective memory task performance has focused on overall task performance. As Dobbs and Reeves (1996) point out, the finding that brain injured persons show deficits in prospective memory task performance relative to controls does not lead to the conclusion that prospective memory retrieval (i.e., unprompted cue recognition) is compromised in those people. Performance deficits may have occurred as a result of poor performance in any one of the task components or phases that make up prospective memory tasks. Furthermore, although there is little doubt that a number of the prospective memory task components depend on frontal functions, the picture is less clear with regard to the prospective memory component itself. Is unprompted cue recognition a function of a different set of neural mechanisms or pathways than those responsible for the cued recall that occurs when attempting to remember the significance of the cue? Both the recognition-recall and the automatic associative activation models of prospective memory maintain that the cue recognition that occurs in prospective memory tasks happens reflexively and automatically. Through activity of the hippocampal system, an effective retrieval cue/context will spontaneously and automatically lead to recognition of the cue. Therefore, there would be little input from the frontal lobes (McDaniel et al., 1999). However, in his functional account of memory Craik (1986) argued that prospective memory tasks provide little environmental support and therefore place a heavy demand on self-initiated retrieval processes. In fact, he proposed that because there is no overt external prompt to remember in prospective memory tasks, they depend on self-initiated processing to a greater extent than other memory tasks, such as free recall, cued recall, and recognition. Thus, according to this account of prospective remembering, the prospective component (unprompted cue recognition) of prospective memory tasks should rely substantially on frontal lobe functioning. Functional Incompatibility And Prospective Memory Functional incompatibility exists when the function of one memory system cannot, because of it's specialized nature, effectively serve functions of other memory systems (Sherry & Schacter, 1987). Is it the case that the unique specializations attributed to episodic memory do not include the ability to remember previously formed intentions? Conversely, is prospective remembering specialized to such an extent that it evolved separately from episodic retrospective memory and cannot control the recollection of past experiences? As discussed previously, one way in which memory 83 systems are proposed to be specialized is in terms of the nature of the "stored" information. Episodic memory is concerned with autobiographical knowledge. It appears that most prospective remembering involves recollecting autobiographical information as well. When one encounters a prospective retrieval cue and recognizes it as a sign of an intention or plan, that recognition is accompanied by the awareness that the plan was formed earlier (see Graf & Uttl, 1999). Many prospective memories are associated with a definite time and/or place of occurrence in one's personal life. Of course, as mentioned earlier, some prospective memory tasks are habitual, and these tasks would be more likely to involve semantic or procedural knowledge (Wilkins & Baddeley, 1978). Nevertheless, the type of prospective memory tasks that are of interest here do seem to be episodic in nature. A second way in which retrospective and prospective remembering may be uniquely specialized is in terms of their temporal focus. Retrospective memories are about the past, whereas prospective memories are about the future. However, as noted earlier, retrospective memories can be about the future and prospective memory retrieval is oriented toward the past. That is, prospective memory cue recognition requires a recognition of something from the past. All recollection, by definition, is aimed at the past. It is intention that is directed toward the future. A fascinating ability of the human mind is the capacity to mentally travel through time. Our capability of experiencing extended subjective time enables us to relive experiences in the past and to mentally project ourselves into the future. Tulving and colleagues (e.g., Tulving, 1985, 1993, 1999a; Wheeler, Stuss, & Tulving, 1997) have labeled this ability autonoetic consciousness. Autonoetic consciousness enables us to mentally represent and become aware of subjective experiences in the past, present, and the future. According to Tulving, it is the hallmark of episodic memory. The episodic 84 memory system not only allows individuals to recollect their past, but also mediates their awareness of their personal future. Through episodic memory and autonoetic consciousness, we use information from our personal past and present to create intentions, set goals, and plan actions for our personal future. Thus, it appears likely, at least in terms of content and temporal focus, that retrospective and prospective memory retrieval are not functionally incompatible, but, instead, are likely to be governed by the same memory "system" (in this case, episodic memory). 85 CHAPTER SIX Aging and Prospective Memory: Is There a Developmental Dissociation Between Prospective and Retrospective Remembering? 86 Retrospective and Prospective Memory: The Effects of Age A great deal of research has been devoted to the topic of age-related changes in retrospective memory and, in general, the overall pattern appears to be one of decline (Light, 1991). However, the degree of change in memory performance that accompanies age tends to vary with the type of task (e.g., Craik, 1986, 1992, 1994). For example, older adults have a tendency to perform more poorly on tasks of free recall than cued recall or recognition (e.g., Craik & Jennings, 1992; Craik & McDowd, 1987). They experience difficulties in memory for temporal information (e.g., Kausler, Lichty, & Davis, 1985; Kausler, Salthouse, & Saults, 1988), memory for source and other types of contextual information (remembering where and when an event took place) (e.g., Burke & Light, 1981; Schacter, Kaszniak, Kihlstrom, & Valdiserri, 1991), and memory for spatial locations (e.g., Ohta & Kirasic, 1983; Uttl & Graf, 1993). Working memory shows a substantial decline with age (e.g., Craik, 1986; Salthouse & Babcock, 1991) as do some aspects of metamemory (e.g., Cavanaugh, 1989; Dobbs & Rule, 1987). By contrast, a number of memory tasks tend to show little or no decline with advancing age, including implicit memory (e.g., Graf, 1990; La Voie & Light, 1994; Light & Singh, 1987), short-term memory (e.g., Delbecq-Derouesne & Beauvios, 1989), recognition memory (e.g., Craik & McDowd, 1987), and memory for well-learned facts and knowledge (e.g., Light & Anderson, 1983). Considerable insight into the ways in which memory operates has been gained through the examination of the impact of aging on memory functioning. As mentioned previously, it is generally assumed that the most convincing way of showing that two memory tasks are functionally distinct or tap different memory systems is to find experimental dissociations (preferably double dissociations) between the tasks. 87 Similar developmental and neuropsychological dissociations are regarded as strengthening the conclusions that can be drawn from functional dissociations alone (see Nyberg & Tulving, 1996). The need for converging evidence of dissociations using a number of different measures of the tasks in question and testing a variety of different populations has been emphasized by many researchers (e.g., Roediger, 1984; Schacter, 1992). By examining the effects of aging on prospective memory task performance and comparing them to age effects observed on different retrospective memory tasks, it is hoped that a better understanding of the relations between the processes underlying these two types of memory tasks will result. Compared to the level of knowledge about age-related changes in retrospective remembering (see Kausler, 1994 for a review), relatively little is known about the effects of age on prospective memory functioning. However, with the recent surge of interest in the study of prospective memory, the research literature on aging and prospective memory is growing. Probably the biggest factor to spark interest in this topic was the fact that the results of a number of preliminary studies were quite mixed. Some research suggested that tests of prospective memory are particularly sensitive to aging (e.g., Cockburn & Smith, 1988) and that a discrepancy between performance on retrospective and prospective tests may be an indicator of early dementia (Huppert & Beardsall, 1993). In contrast, other research found little or no decline in prospective memory functioning with age (e.g., Einstein & McDaniel, 1990; Maylor, 1990). Furthermore, several studies found older adults to be superior to younger adults in performing prospective memory tasks (e.g., Martin, 1986; Moscovitch, 1982; Patton & Meit, 1993; West, 1988). Studies showing an absence of an age effect on prospective memory tasks led many investigators to postulate that prospective remembering is a unique function of memory, distinct from retrospective memory and relatively immune 88 to the effects of aging. Thus, the question of primary interest that emerged from these findings concerns whether prospective memory test performance declines with age, and, if so, whether the declines are comparable to retrospective memory? Declines in cognitive processing resources have been proposed as responsible forage-related declines in memory (e.g., Craik, 1983, 1986; Hasher & Zacks, 1979; Jorm, 1986; Salthouse, 1985, 1988). A number of different metaphors have been used to conceptualize processing resources including a space metaphor in which a general capacity, such as working memory, is assumed to be the essential processing resource (e.g., Baddeley, 1986); an energy metaphor in which resources are described as a general purpose mental "fuel" (e.g., attention) that drives information processing (e.g., Craik & Byrd, 1982); and a time/speed metaphor which defines resources in terms of the speed with which or the time required to complete a task (e.g., Cerella, 1990; Salthouse, 1996). Nevertheless, in general, a processing resource can be defined as "something that exists in limited supply and is responsible for the enhancing or enabling of certain cognitive processes" (Salthouse, 1990, p. 102). Two general assumptions underlie most resource models of cognitive processing: (1) information processing is constrained by the amount of mental resources available at a given moment, and (2) cognitive activities vary in the mental resources they require for maximal performance. Resource accounts of cognitive aging make the additional assumption that adult age differences in cognitive performance are a result of a decline in the available processing resources that occurs with advancing age. Taken together, these assumptions lead to the prediction that the degree of age-related decline on any particular task will depend on the cognitive resources available to the individual at the time and on the specific demands of the task. 89 Building on resource theories of cognitive functioning, Craik's (1986, 1992, 1994) framework for understanding age deficits in memory tasks argues that performance is determined by both external factors (e.g., cues, context) and the type of mental operations required by the task. As outlined earlier, Craik's functional account of memory distinguishes between mental processing that is guided and supported by the environment and processing that is self-initiated and controlled. When environmental support is available, for example in the form of retrieval cues, then memory retrieval will be guided primarily by the cues. However, when little or no environmental support is present, remembering is heavily dependent upon self-initiated processing (e.g., preexisting knowledge). Craik proposed that aging results in a decreased capacity for self-initiated or controlled processing, and as a result, age-related decrements in memory performance are observed to the extent that specific tasks require self-initiated processing. He also proposed that prospective memory tasks depend upon self-initiated processing to a greater extent than most other memory tasks because they require subjects to remember the intention or plan on their own, without any external prompting for remembering (e.g., experimenter). Therefore, based on the widespread view that aging is accompanied by a decline in the pool of resources available for processing information (e.g., Craik, 1983, 1986; Hasher & Zacks, 1979; Jorm, 1986; Salthouse, 1985, 1988), and the assumption that prospective memory tasks are particularly reliant on self-initiated and controlled processing, it follows that performance deficits due to increasing age should be particularly pronounced in tasks of prospective memory (Craik, 1986; Craik & Jennings, 1992). 90 Prospective Memory and Aging: A Meta-Analvsis In an attempt to clarify the inconsistencies in the research and determine the extent to which prospective memory declines with age, a meta-analytic review of the available research on age differences in prospective memory task performance was conducted (see Birt, 1999). Specifically, the primary theoretical claims about prospective memory and aging were tested, and various subject- and task-related variables, shown to influence retrospective memory performance, were explored as potential moderator variables. The data of 25 studies and 2,695 participants, resulting in 96 effect sizes computed for young - old prospective memory performance differences revealed a large amount of variation among individual study findings. These results substantiate the view that, although relatively new, research in prospective memory is marked by a substantial amount of inconsistency and variability (see Brandimonte, Einstein, & McDaniel, 1996 and Ellis & Kvavilashvili, 2000 for reviews). The results revealed, in general, prospective memory does indeed decline with age. Older adults' scores on laboratory tests of prospective memory ranged between 0.5 and 1.0 standard deviations lower than younger adults' scores. These findings are comparable to those from a meta-analysis of age differences in performance on episodic retrospective memory tasks which found that older adults performed an average of 0.7 to 1.0 standard deviations lower than younger adults (Verhaeghen, Marcoen, & Goossens, 1993). Consistent with the pattern of age-related changes in retrospective memory performance, the degree of age-related change observed in prospective memory tasks varied greatly as a function of the specific features of the task. For example, one of the most important factors found to influence the magnitude of age differences in 91 prospective memory task performance was the type of experiment. Artificial laboratory experiments (most often employing computer tasks) resulted in the largest performance decrements (d = .74) followed by naturalistic laboratory experiments (d = .55), which tend to utilize more natural, everyday kinds of prospective memory tasks within a laboratory setting. However, naturalistic field studies, which test memory functioning in real world settings, often showed a performance advantage for older adults relative to younger adults (d = -.18) (see also Rendell & Thompson, 1999). Age differences also were dependent on whether the prospective memory task employed an event-based or time-based retrieval cue. It was found that compared to younger adults, when task performance is tested in a controlled laboratory setting, older adults perform more poorly on time-based (d = .99) than event-based (d = .70) tasks. However, when naturalistic field studies are conducted and prospective performance is tested in the real world (where they are free to use memory aids), older adults perform much better on time-based tasks, such as showing up for appointments promptly, than do younger adults (d = -.60). A number of other variables also were found to influence age differences in performance on prospective memory tasks, including age of the older adults, ongoing task load, total number of trials, length of the retention interval, the amount of practice with/ exposure to the retrieval cue, the size of the response window, and the scoring criterion (see Birt, 1999 for details). Thus, the results from the meta-analysis indicate that in the absence of external memory aids or reminders, prospective memory task performance tends to decline with age. Consistent with the experimental literature on aging and retrospective memory tasks, age-related changes in prospective memory performance are dependent on task characteristics and therefore are highly variable. Consequently, the findings associated with aging populations appear to indicate that prospective and episodic 92 retrospective memory task performance are similarly influenced by advancing age, which lends further support to the notion that they may not represent separate memory functions or systems. With respect to the quantitative review carried out to examine the patterns of prospective-retrospective memory task dissociations in the research literature, 12 of the 44 studies reviewed examined age differences in both prospective and retrospective tests of memory (see Appendix A). Out of a total of 25 prospective-retrospective test comparisons included in these studies, 2 showed a double dissociation (age x task load interactions), 10 were single dissociations, and 13 failed to show dissociations. Recall, recognition, and short-term memory tests did not differ in frequency of dissociating with prospective tests. This pattern of results appears to contradict Craik's (1986) prediction that age-related performance decrements should be especially pronounced on prospective memory tasks due to a strong reliance on self-initiated processing. These results also appear to be inconsistent with the recognition-recall and the automatic associative activation models of prospective memory. Because both models maintain that prospective remembering is primarily spontaneous and automatic, one would expect to see little or no decline in performance with increasing age. Nonetheless, again, it is important to keep in mind that these results are largely based on measurements of overall prospective memory task performance (and often very different prospective and retrospective memory tasks). Although some studies included in Appendix A and in the age difference meta-analysis attempted to isolate the prospective memory task component by either minimizing the influence of the retrospective component or by using lenient scoring criteria, the extent to which prospective memory retrieval declines with age and how the degree of decline compares with those observed in different retrospective memory tasks remains to be determined. CHAPTER SEVEN Self Assessment of and Individual Differences in Prospective Memory Functioning 95 As the quantitative review of the patterns of prospective-retrospective memory performance dissociations and the meta-analysis of age differences in prospective memory performance revealed, there tends to be a great deal of variability in performance on prospective memory tasks. These reviews also revealed a number of task/ methodological variables associated with that variability, such as type of experiment (artificial laboratory, naturalistic laboratory, and naturalistic field), type of retrieval cue (event- vs. time-based), amount of practice with retrieval cues, ongoing task load, retention interval, response window, scoring criterion, etc. Recently, researchers have recognized the need to investigate individual differences in personality, cognitive abilities, and meta-cognitive processes in order to construct a more complete picture of the factors associated with prospective remembering and to more fully understand the sources of the variability in prospective memory retrieval (e.g., McDaniel & Einstein, 2000b). To the extent that prospective memory represents a unique function or activity of memory, it seems likely that individual difference factors can be identified that are more important for prospective than for retrospective memory retrieval (Meacham, 1988; Winograd, 1988). Self Assessment Of Prospective Memory Performance Rather than focusing on actual memory performance, self-report measures of memory assess individuals' appraisals and perceptions of their memories. Knowledge, perceptions, and beliefs about one's own memory functioning and the functioning of memory in general is referred to as metamemory (e.g., Dixon, 1989; Hertzog & Hultsch, 2000). Metamemory also includes knowledge of the demands of different memory tasks and an understanding of what strategies might improve memory performance. In general, there appears to be little relationship between beliefs about 96 memory and actual memory performance (see Herrmann, 1984; Kausler, 1994; for reviews). However, when there is a close match between subjective and objective measures of memory and when assessment tools with good psychometric properties are used, significant positive correlations can result (e.g., Dixon & Hultsch, 1983; Hertzog et al., 2000; Sunderland, Watts, Baddeley, & Harris, 1986). The study of metamemory in the elderly is quite important because older adults' beliefs about their memorial capabilities may influence the types of memory-demanding tasks they attempt, their actual performance on memory tasks, and the memory strategies they employ to aid task performance (Cavanaugh, 1989). Prior research investigating age differences in self-reports of memory functioning has yielded mixed results. Whereas some studies have found that older adults give lower ratings of their memory abilities than do younger adults (e.g., Dixon & Hultsch, 1983), others have found older adults' ratings to be higher than those of younger adults (e.g., Bennet-Levy & Powell, 1980). Although few subjective memory or metamemory questionnaires have focused primarily on prospective memory performance (cf. Hannon et al., 1995), a number of questionnaires include items that can be classified as prospective memory tasks (e.g., remembering appointments, paying bills, taking medication). Maylor (1993b) reviewed the literature on aging and self-assessments of prospective memory task performance and found that, in general, there appears to be a positive relationship (i.e., ratings of good prospective memory performance tend to increase with age). In fact, a number of studies have found opposite ratings of prospective and retrospective memory task performance with age. For example, Cohen and Faulkner (1984) and Martin (1986) found that although older adults gave high ratings for prospective memory questionnaire items, they tended to give much lower ratings for a number of retrospective memory items. Maylor (1993b) suggested 97 several reasons why older adults perceive prospective memory functioning to be relatively more preserved than retrospective memory functioning in old age. First, regardless of age, individuals may not be very good at judging the proficiency of their own memories. For example, there are likely many instances in which one does not become aware of having forgotten to do something (Wilkins & Baddeley, 1978). Second, older adults may use compensatory strategies such as external cues or reminders to aid remembering. There is evidence to suggest that the use of external memory aids increases with age, at least with some tasks (e.g., Lovelace & Twohig, 1990; Maylor, 1996b; Reeves & Dobbs, 1992). Furthermore, the ability to use effective strategies may increase with age and experience (Martin, 1986). Third, younger and older adults may differ in their number of opportunities to forget. For example, if compared to younger adults, many older adults lead a more relaxed life style, are less busy, or do not make as many plans and commitments, then they have fewer opportunities to forget them. One goal of the present research was to explore further the relationship between self-assessments of prospective remembering and actual performance on tests of prospective memory retrieval. Using a questionnaire designed specifically to provide self-assessments of prospective memory performance (Hannon et al., 1995), the present study investigated whether meta-cognitive perceptions and evaluations of prospective memory ability correspond with demonstrated capabilities on two different tests of prospective memory retrieval and explored how this relationship might differ as a function of age. 98 Individual Differences In Prospective Memory Performance Only a handful of studies have examined the relationship between individual differences and prospective memory task performance (e.g., Cherry & LeCompte, 1999; Cockburn & Smith, 1991, 1994; Harris & Menzies, 1999; Mantyla & Nilsson, 1997; Rude et al., 1999; Searleman & Gaydusek, 1989; Wichman & Oyasato, 1983). Some of these studies focused primarily on differences in cognitive abilities and others focused on differences in personality or mental states. Cockburn and Smith (1991) examined the relationship between prospective memory and a variety of individual difference measures, including fluid intelligence (as measured by Raven's Coloured Progressive Matrices; Raven, 1984), crystallized intelligence (as measured by the National Adult Reading Test; Nelson, 1982), age, years of education, participation in social activities, and health status. Using the three prospective memory items on the Rivermead Behavioural Memory Test (RBMT; Wilson, Cockburn, & Baddeley, 1985) as an index of prospective memory ability, they found that only two of the individual difference measures (fluid intelligence and age) accounted for a significant proportion of variability in task performance. As a part of a large-scale population based study of aging, memory, and health, Mantyla and Nilsson (1997) examined the relationship between performance on an event-based naturalistic laboratory prospective memory task and a number of factors, including age, word fluency, word comprehension, block design and mental status (as measured by the Mini-Mental State Examination, M M S E ; Folstein, Folstein, & McHugh, 1975). Both age and word fluency significantly predicted prospective memory task performance. Cherry and LeCompte (1999) compared two samples of younger and older adults that differed in their level of educational attainment, occupational status, and verbal ability (measured by the Wechsler Adult 99 Intelligence Scale, WAIS; Wechsler, 1955). They found that the younger adults and the higher ability older adults (i.e., those with high levels of education, occupational status, and verbal ability) performed better on a computerized laboratory-based test of prospective memory than did the lower ability adults. In addition, working memory span and recognition test performance accounted for a significant, albeit small, proportion of the variance. Uttl et al. (1999) found only modest correlations between naturalistic laboratory tests of prospective memory and a variety of cognitive tests, including digit symbol, card sorting, verbal fluency, picture-word Stroop, and visual search. Personality/mental states such as anxiety, stress, depression, and locus of control also have been investigated in relation to prospective memory task performance. Anxiety level has been associated with decrements in performance on measures of retrospective memory in many studies (see Eysenck, 1981). Likewise, Harris and Menzies (1999) found a negative relationship between level of anxiety and prospective memory performance. However, Cockburn and Smith (1994) found the relationship between anxiety and performance on prospective memory tasks (on the RBMT) to be more complex . They found that both low and high levels of anxiety can facilitate prospective memory task performance, but that intermediate levels have detrimental effects. Although self-reported depression has been found to be associated with decreased performance on a number of retrospective memory tests (e.g., Hertel & Rude, 1991), Harris and Menzies (1999) found no relationship between level of depression (as measured by the Depression Anxiety Stress Scale, D A S S ; Lovibond & Lovibond, 1995) and performance on an event-based laboratory prospective memory task. By contrast, Rude et al. (1999) found depression-related impairments on a time-based prospective memory laboratory task (using the Beck 100 Depression Inventory, BDI; Beck & Steer, 1987; and a Structured Clinical Interview, SCID; Spitzer, Williams, & Gibbon, 1987), and Hannon et al. (1995) found that lower scores on the BDI (indicating fewer symptoms of depression) were associated with better performance on a variety of event- and time-based prospective memory tasks. Several personality factors have been found to be related to prospective memory ability as well. For instance, Searleman and Gayduseuk (1989) reported that people with Type A personalities, who tend to be highly conscientious, perfectionistic, tense, competitive, and overly concerned with deadlines, are more likely to remember to perform prospective memory tasks and perform them quickly than people who do not have such personality characteristics. Also, Wichman and Oyasato (1983) found that whereas persons with external locus of control (perceive outcomes as due to factors independent of their behaviour) tend to perform better on prospective memory tasks that are simple and repetitious, persons with internal locus of control (interpret outcomes as contingent on their own behaviour) show better performance when task complexity and difficulty increases. If it is the case that unprompted cue recognition is a key element in what makes prospective remembering unique, then one would expect factors relating to the ability to notice and recognize cues embedded in the retrieval environment to be significantly related to prospective memory task performance. As indicated above, a number of individual difference factors found to be associated with measures of retrospective memory have shown similar associations with measures of prospective memory (e.g., verbal fluency, ability level, anxiety, depression). Are there any personality or cognitive ability factors that might be associated with prospective memory task performance to a greater extent than retrospective performance? Given that scores on measures of verbal ability and intelligence have been associated with prospective memory test 101 scores and given that the prospective memory tests and targets in this study are of a verbal nature (i.e., words), this relation was examined more closely. With respect to unprompted cue recognition, it seems likely that individuals who are better at noticing cues in the environment and picking out stimuli embedded in an array of familiar and similar stimuli (e.g., visual search) would be better at prospective memory retrieval (especially when the prospective memory cues are primarily visual). Correspondingly, persons who have a tendency to dissociate (i.e., separate attention, thoughts, and experiences from consciousness) in daily life might be expected to fail to notice many of the prospective memory cues that present themselves in their environments due to a the tendency to detach from reality and a high susceptibility to distraction. Dissociation has been shown to be strongly associated with memory performance in a variety of contexts, including the creation of false memories (e.g., Hyman & Billings, 1998; Porter, Birt, Yuille, & Lehman, 2000) and memory for traumatic events (e.g., Porter & Birt, 2000). It is possible that dissociation may be related to prospective memory test performance, especially with respect to ability/ tendency for noticing retrieval cues. Given that most "longer-term" prospective memory tasks include an activity-filled retention interval that exceeds short-term memory capacity, carrying out the intention is delayed and opportunity to rehearse the intention is reduced. Nevertheless, it seems likely that some individuals may be more prone to experiencing spontaneous intention-related thoughts during the retention interval. Associated with the notion of involuntary or spontaneous thought is the form of mental control referred to as thought suppression—a deliberate effort to not think about a particular thought. A substantial body of literature exists demonstrating the paradoxical nature of thought suppression, indicating that attempts to suppress thoughts often make those thoughts more accessible and intrusive (see Wenzlaff & Wegner, 2000 for a review). This "rebound 102 effect" has been found with emotional thoughts such as the content of a distressing film (Davies & Clark, 1998) as well as neutral thoughts such as a 'white bear' (Wegner, Schneider, Carter, & White, 1987) or a 'green rabbit' (Clark, Ball, & Pape, 1991), and one explanation for its occurrence is that it stems from the motivation to fulfill a blocked goal. This idea is founded in what has been labeled the Zeigarnik effect—the motivated persistence of goal-related thoughts (Lewin, 1951). Zeigarnik (1938) found that individuals had better recall of an experimental task if they had not completed it due to an interruption. This effect has been taken as evidence that thoughts related to an incomplete task remain mentally active, and because the original goal is unfulfilled, intermittent thought intrusions are likely to occur during and after suppression (and in the case of a prospective memory task, during and after the retention interval) (see Goschke & Kuhl, 1996; Mantyla, 1996). Wegner and Zanakos (1994) suggested that there may be individuals who show a general tendency to use thought suppression across a variety of situations and thought topics and devised a self-report index, called the White Bear Suppression Inventory (WBSI), to measure such a tendency. It is conceivable that individuals scoring low on the WBSI (i.e., those with higher frequencies of intrusive thoughts) may perform well on prospective memory tasks given the possibility of heightened intention-related thoughts during the testing session. To summarize, in an attempt to identify individual difference factors related to performance on prospective memory tests, verbal ability/ intelligence, visual search, dissociation, and thought suppression tendencies were assessed in relation to objective prospective memory task performance, self-reported prospective memory ability, as well as retrospective memory performance. Potential relations with age also were explored. CHAPTER EIGHT The Research Study 104 Summary of Study Objectives The primary purpose of this research study was to compare the similarities and differences between prospective and retrospective memory retrieval to test the hypothesis that prospective memory is a functionally distinct activity of memory. The goal was not necessarily to provide evidence that prospective memory is a distinct, dissociable new memory "system," but rather to focus on the memory demands of explicit, episodic prospective tasks and uncover how they are related to and unique from those in various retrospective memory tasks. The results of a quantitative review of the patterns of dissociations within the existing research literature revealed almost as many instances in which task and subject variables produced similar effects on prospective and retrospective memory task performance as different effects. However, the pattern of results from stochastic dissociations provided some evidence to suggest independence. Performance on retrospective recognition tests was somewhat less likely to correlate with performance on prospective memory tests than performance on retrospective free recall tests. Based on the failure to find correlations between prospective and retrospective remembering, a number of studies have concluded that these are different forms or functions of memory. However, such studies often have compared prospective and retrospective memory test performance while using very different tests to measure each (e.g., medication adherence and free recall; Wilkins & Baddeley, 1978). By contrast, the goal of the present study was to examine the relationship between prospective and retrospective memory tests that were equated on as many task features as possible. The purpose of matching the test properties (e.g., cues, study time, retention interval, response type, context, environment) as closely as possible, was to minimize the influence of potential confounds and facilitate the 105 isolation of any specific mental processes that may fundamentally differentiate prospective from retrospective memory retrieval. It also allowed a much more precise measurement of the extent to which such retrieval processes are related, thereby addressing the question of whether they are functionally distinct. In order to gather further evidence for this question two different age groups (young and older adults) were tested and a levels of processing manipulation was incorporated, thereby making it possible to test for developmental and experimental dissociations between retrieval on the prospective and retrospective memory tasks. A second, and closely related, purpose of this research was to examine age differences in prospective memory retrieval. Craik's (1986) functional model of memory argues that the memory processes underlying recognition of the prospective retrieval cue as a sign of a previously formed intention are largely self-initiated and effortful. Based on the argument that the pool of processing resources available for carrying out mental activity decreases with age and the assumption that prospective memory retrieval is heavily dependent on self-initiated processes, Craik (1986) predicted that age-related decrements in performance on prospective memory tests should be more pronounced than those observed on retrospective memory tests. Other models of prospective memory, for instance, the recognition-recall model and the automatic associative activation model, argue that unprompted cue recognition occurs automatically and predict smaller age changes in prospective remembering than those predicted by the functional account. A meta-analysis of the age difference effect sizes in performance on prospective memory tasks within the research literature was conducted to test such theoretical claims about prospective memory and aging as well as to explore a number of potential moderator variables (see Birt, 1999). Overall, it was found that prospective memory retrieval does appear to decline with age and the 106 degree of decline was comparable to that found for episodic retrospective memory retrieval. However, the size of the age differences in prospective memory performance was found to vary greatly as a function of a many different task features. In an attempt to reduce such variability and obtain a clearer picture of the relationship between prospective and retrospective memory retrieval as well as the effects of age, a research study was conducted in which the scope of inquiry was narrowed to episodic prospective memory retrieval and as many of the potentially confounding task factors were controlled for. In order to ascertain how prospective memory retrieval fits into the current understanding and conceptualizations of memory retrieval, three different types of tasks were administered to test retrospective memory: two explicit, episodic tests (cued word recall and word recognition) and one implicit test (word completion). As stated previously, explicit tests of memory, such as recall and recognition tests, require participants to recollect information from a previous study episode in a conscious, deliberate manner. By contrast, tests of implicit memory, such as word stem (str ) or word fragment completion (s _ r t), measure the influence of a prior study episode in the absence of any conscious, intentional attempt at recollecting that episode. Explicit and implicit memory tasks have been demonstrated to be functionally distinct and dissociable tests of retrospective memory, measuring different ways of accessing memory for previous events or episodes (e.g., Graf & Schacter, 1985). However, as outlined above, it is not known whether prospective memory represents yet another functionally distinct aspect of memory. Some researchers have noted that there seems to be a spontaneous remembering component in prospective memory tasks and that the processes underlying prospective remembering appear to be largely automatic (e.g., Einstein & McDaniel, 1996; Goschke & Kuhl, 1996). These 107 observations have led to the speculation that prospective remembering may be more closely related to implicit rather than explicit tests of memory (e.g., Einstein & McDaniel, 1996). To help resolve this debate, a direct, precise comparison (achieved by matching test properties as closely as possible) between these types of memory tests was conducted. Further, because most explicit memory tests are affected by variations in depth of processing at encoding and most implicit tests are not, a levels of processing manipulation was employed during the initial study phase. Specifically, it was hypothesized that if the processing demands of prospective remembering more closely resemble the memorial processing that occurs in explicit tests such as cued recall and recognition, task performance should be influenced by the levels manipulation. However, if prospective memory retrieval is more similar to implicit retrieval, performance should be relatively insensitive to depth of processing. Three tests of prospective memory were employed in this study: two objective tests and one subjective test. One of the objective tests utilized word stems (e.g., res ), word fragments (e.g., res_d c_), and word beginnings (e.g., residence) as the prospective memory cues presented in the context of the implicit word completion and the explicit cued recall and word recognition tests. The second prospective memory test embedded prospective cues (words beginning with 'res', such as residence) in a story reading task. The subjective prospective memory test was the Prospective Memory Questionnaire (PMQ; Hannon et al., 1995), a self-rating questionnaire that assesses perceptions of prospective memory task performance in everyday life. As stated above, creating measures for testing prospective memory performance has presented somewhat of a challenge. A number of different methods have been used to date, and most are assumed to be measuring the same mental activity—prospective remembering. However, different tests of prospective memory 108 have shown different patterns of memory performance. Another goal of the present study was to compare performance on different prospective memory tests specifically designed to be similar in many ways (e.g., cues, study time, response type, environment, etc.), but that employ different methodologies. The purpose of this comparison was to: (1) examine the degree to which these different prospective memory tests measure the same memory processes, and (2) potentially identify and/or rule out possible test characteristics resulting in performance differences. Finally, given the considerable variability in performance on prospective memory tasks that has been demonstrated in the research, another important purpose of this study was to explore several individual difference factors in an attempt to identify potential sources of such variability in performance. Given that the ability to notice cues in the environment as reminders of previously formed intentions appears to be critical for success in prospective remembering, factors thought to be related to this ability were explored (e.g., visual search, thought suppression, and dissociative tendencies). To summarize, the present research study included implicit retrospective, explicit retrospective, and prospective tests of memory-all equated on as many task dimensions as possible, while still maintaining the integrity of the individual tests. The prospective memory tests were limited to those of an episodic nature. In addition, only event-based prospective memory tasks were investigated. Although aspects of the theoretical and empirical work presented may also apply to time-based tasks, they were not explicitly considered. Using a within-subjects design, the relationship between performance on these tests was examined in detail. In addition, the performance of young adults on these tests was compared to that of older adults, in an attempt to gain further insight into the relations between these tests, to explore the pattern of age-related differences in 109 performance on the tests, and to increase understanding of how memory changes with age. Finally, and most importantly, because research in prospective memory is relatively new, it was hoped that this research would provide much needed insight into how it is we remember to do things in the future. Method Participants A total of 132 participants (66 young and 66 older adults) were included in this study. The younger adults (M = 20.09 years; SD = 3.35; range = 17-39) were undergraduate students recruited through the Dalhousie University Psychology Department's student subject pool and received course credit for their participation. The older participants (M = 75.42 years; SD = 7.32; range = 63-92) were healthy community-dwelling adults over the age of 60, most of whom were recruited via sign-up sheets and advertisements circulated at volunteer programs at a local geriatric hospital (Camphill Veteran's Hospital), elderobics classes at local community centers, as well as various church groups. Twenty-two of the older participants were recruited from the Halifax Center component of the Canadian Study of Health and Aging (CSHA). The C S H A is an ongoing, longitudinal nationwide investigation of the health and cognitive status of persons age 65 years and older in Canada (see Canadian Study of Health and Aging Working Group, 1994). Community-dwelling individuals who scored > 85 on the Modified Mini-Mental State Examination (3MS; Teng & Chui, 1987) during the most recent phase of testing (third phase) were contacted by telephone and invited to participate subsequent to their involvement in the CHSA. A total of 46 individuals were contacted, 8 of whom did not participate because of health reasons, 5 reported being too busy to participate, and 11 were not interested in participating in the study. With 110 the exception of 3MS scores, all data for the 22 participants recruited through the C S H A reported in this study were collected independently of the C S H A testing. Those participants not recruited through the C S H A were administered the 3MS at the time of testing, and all scored above the cut-off score of 85. Participation for all older adults was strictly on a volunteer basis. General descriptive data on the participants are shown in Table 2. All participants completed a demographics questionnaire (see Appendix D) that included questions assessing age, gender, ethnic background, educational attainment, occupational status, social and activity characteristics, and several perceived health questions. With respect to gender, although the ratio of number of males to females in the younger and older age groups was the same, there were many more female than male participants. The vast majority of participants in both age groups were Caucasian whose native language was English. As they were recruited within a university setting, all of the younger participants obtained at least some university education with 60.6% enrolled in either their first or second undergraduate year at the time of testing. Educational level was more varied for the older participants, nevertheless 65.1% did achieve at least partial college or some form of specialized training. For the occupational status question, the younger adults were asked to rate the job status of their same sex-parent (see Cherry & LeCompte, 1999), and 87.9% of parents were rated as skilled or higher (with 40.9% at the professional level). Similarly, the majority of the older participants (89.4%) had a pre-retirement occupational status of skilled or higher (with 33.3% at the professional level). Older participants reported involvement in a greater number of social clubs and organizations than did younger participants. However, younger participants reported spending more hours outside their homes than did older participants. 111 As indicated in Table 2, there was no significant difference in self-rated general health between the young and older groups. When asked how their health compared to other people their age, older subjects' comparison ratings were somewhat higher than those of the younger subjects, and when questioned about the extent to which their health interferes with participating in activities, older adults reported somewhat more interference than the younger adults. Self-ratings of quality of eyesight and hearing fell primarily within the excellent to good range for both age groups, although older participants indicated slightly poorer quality than younger participants. Only the older participants were administered the 3MS, and all scores were well within the normal range (86-100; see Table 2 and description below). Materials and Design The main design of this study comprised a mixed factorial with four factors (see Table 3). The between-subjects factor was age (young and old). The within-subjects factors were study orienting task (semantic and nonsemantic), retrospective memory test (word completion, cued recall, and recognition), and prospective memory test (prospective memory test 1 and prospective memory test 2). Self-perceptions of prospective memory functioning also were examined for all participants. In addition, visual search ability, dissociative tendency, thought suppression, verbal ability/ intelligence were explored in relation to both subjective and objective prospective memory test scores. Assessment of cognitive functioning. In order to ensure that the older participants included in this study were cognitively healthy, global cognition was assessed by the Modified Mini-Mental State Examination (3MS; Teng & Chui, 1987) where scores can range from 0-100. Higher scores indicate better performance. The 112 3MS is a modification of the Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975). Specifically, four items were added, the scoring system was refined, and clearer instructions for scoring were given (Teng & Chui, 1987). Compared to the M M S E the 3MS has a better sensitivity and specificity for detecting dementia (McDowell, Kristjansson, Hill, & Hebert, 1997). The 3MS can be administered within 10-20 minutes and requires no special equipment. A score of less than 78 is widely used as the cut-off point for the possibility of cognitive impairment (see Canadian Study of Health and Aging Working Group, 1994). To increase the sensitivity of the test even further, a cut-off score of 85 was used in the present study. Stimulus materials. The materials selected for the memory tests included in this study were 51 words containing an average of 6.39 letters (Mode = 6), with a range of 4 to 9 letters (SD = 1.17). Most of the words were taken from the Graf and Williams (1987) word stem completion norms. Each word begins with a different three-letter stem for which Webster's (1988) compact dictionary gives at least 10 alternatives beginning with that stem (M = 31.69; SD = 14.88). Words were selected such that they had a Thorndike and Lorge (1944) frequency ranging from 8 to 50 (M = 29.61; SD = 13.81) and a Kucera and Francis (1967) frequency ranging from 4 to 48 (M_ = 19.75; SD = 11.13), with none being the most frequent or infrequent completion for its stem. The words also had a mean frequency of occurrence as a first word stem completion of 2.36% (Mode = 2), with a range of 1-6% (SD = 1.21) (Graf and Williams, 1987). An additional 6 words with similar properties were selected for use as practice items. Each word was typed in lowercase letters in Arial 24-pt. font on separate 4 x 6 index cards. Examples of the stimulus words are "triumph", "deputy", and "weapon" (see Appendix B for the complete list of words). 113 Of the 51 primary stimulus words, 48 were divided into two sets of 24, List A and List B, for use as alternative study lists. Three words were reserved for use as prospective memory test targets (i.e., words beginning with the letters "res"). Half the subjects in each age group studied List A, half studied List B. Within each set, the words were divided into two equal subsets comprised of 12 words each, one for presentation in each of the two study orienting tasks. These four subset lists were balanced for word length, word frequency, number of available completions, and first completion frequency. Within both study List A and B, each subset list was used equally often for the semantic and nonsemantic tasks. These orienting tasks were presented to participants in random order. Two of the three prospective memory target words were added to each of the study lists. One was studied semantically and one was studied nonsemantically. The third target word served as a nonstudied control. The assignment of each of the three prospective memory target words to study conditions was counterbalanced. Levels of processing manipulation: The rating & counting task. The "rating and counting task" represented the study phase for the memory tests. For this task, participants were asked to work with a total of 26 words (either List A or List B plus two prospective memory target words) presented on cue cards one at a time. Each word was shown for approximately five seconds (subject-paced), the participants read the words aloud, and a levels of processing manipulation was employed. Half of the words were accompanied by instructions that focused processing on the semantic components of the words (pleasantness rating), whereas the other half directed attention and processing to the surface features of the words (consonant counting). For the pleasantness rating study orientation (semantic/ deep encoding) participants 114 were asked to "look at each word carefully as it is presented, to think about its meaning, and to decide how pleasant or unpleasant you feel the meaning of the word is to you." Consistency in the ratings was emphasized by instructing participants to "think about each word and rate it in such a way that you would be able to give a very similar rating if you saw the word again." Each item was rated according to a 5-point Likert-type scale (1 = very pleasant, 5 = very unpleasant). For the consonant counting study orientation (nonsemantic/ shallow encoding) participants were directed to "count the number of consonants in each word as it is presented." Again, half of the words were assigned the pleasantness rating task and half the consonant counting task. Order of presentation was random for each participant. All responses were written by the participant in a test booklet that was provided at the beginning of the task (see Appendix D). It is important to note that the levels of processing manipulation applied primarily to the retrospective tests of memory. For these tests, each participant studied 13 target words semantically and 13 targets nonsemantically. Because there were only two target words in each study list that began with the letters "res", the levels manipulation for the prospective memory tests occurred only with a total of two words per participant. Thus, inclusion of different levels of processing at study served primarily as a manipulation check for the implicit and explicit retrospective memory tests, and tests for a levels effect in the prospective memory task performance were exploratory in nature. Retrospective memory tests. For the retrospective memory tests, there were two different test lists, each consisting of a total of 30 different items. Half of the items were three letter word stems and half were word fragments. The word fragments were 115 designed in such a way that the first three letters of the cue word were always intact (e.g., res_d_n ). This was necessary because the prospective memory cue was the three-letter combination "res" at the beginning of a word, and in order to match test cues as closely as possible across the various tasks, the fragment completion items had to be constructed in that way. For each test list, 27 of the stems/fragments were from the original lists of target words, and three served as fillers. The fillers were located at the beginning, the middle, and the end of the test list. Within each test, 12 of the word stems and fragments were from study List A, 12 were from study List B, and 3 were prospective memory targets (2 studied, 1 nonstudied; note that the three targets were included in both tests). Thus, for any one participant the test was divided into an almost equal number of studied and nonstudied items from the original word list. The nonstudied items provided the baseline measure of priming. Furthermore, all of the 14 previously studied items in each test were divided equally between those presented in the semantic task and those presented in the nonsemantic task. Participants in each group completed the two different word completion tests and so were tested for all 51 stimulus words. The test list used for word stem/fragment completion for half the participants in each group was used for cued recall for the other half, and vice versa. Thus, for both study and test lists, materials were fully counterbalanced across all other experimental conditions. However, all participants carried out the word completion test before the cued recall test. In addition, the word recognition test always followed the cued recall test for every participant. There were two alternative recognition test forms, one for each study list. Besides the original target words, another 96 words were selected as distractors for the recognition test. Each test form was made up of the 24 words from one study list and 48 distractors along with the three prospective memory target words (2 studied, 1 nonstudied). The 116 distractors were chosen such that for each list word there were two distractors that shared the same stem. For example, the target list word "triumph" had the distractors "tribute" and "trivial". Furthermore, all distractors had the same number of letters as the target list words. The list words (with the exception of the prospective memory targets) and the distractors were randomly arranged on the test forms and all words were typed in lowercase letters. Prospective memory tests. The first prospective memory test occurred in the context of the word completion, cued recall, and word recognition tests. The cue for this test was the three letters "res" at the beginning of a word. This cue was included in the form of a word stem and a word fragment in both the word completion and the cued recall tests. For both the completion and cued recall tests, it occurred three times (in the 7 t h , 14 t h , and 2 3 r d positions). Because word fragment tests are far more constrained in the number of possible completions than word stem tests, and to decrease the possibility of overlap in completion responses, the word fragments were always presented before the word stems in both tests. Participants were instructed not to repeat any answers. In addition, the three target words beginning with the letters "res" were included in the word recognition test to serve as prospective memory retrieval cues ("rescue", "residence", and "restless"). These cue words occurred in the 20th, 40th, and 60th positions in both forms of the recognition test. All other words were randomly organized. Thus, a total of 9 prospective memory cues were presented in the first prospective memory test: 3 in the implicit word stem/fragment completion test, 3 in the cued recall test, and 3 in the word recognition test. The second prospective memory test, referred to as the reading task, was modeled after a similar test devised by Kvavilashvili (1998). For this task, the short 117 story "/Agnes" by Gregory Clark was used (King, LeDrew, & Porter, 1982; see Appendix E). This story was modified slightly for the purposes of this study, and comprised four double-spaced typed pages with a total of 841 words. It takes approximately six minutes to read aloud. Similar to the first prospective memory task, the prospective memory cue for this task was the letter combination "res" at the beginning of a word. A total of 9 cue words were included in the story. They were spaced an average of 98 words apart (ranging from 90 to 121 words) and occurred twice on every page, except for the second page which included three cue words. The cue words (i.e., words beginning with the letters "res") included in the story were: "resentfully," "resemblance," "resting," "resist," "reserve," "restaurant," "resonating," "respiratory," "responded." This test was followed by a brief test of story comprehension, questions measuring subjective task difficulty, interest, absorption and willingness to comply, a brief test of memory for the prospective cue (i.e., the retrospective component), and several questions concerning subjective accuracy of performance on the task (see Appendix D). It is worth noting that repeated-intention designs were purposely selected for both prospective memory tests for several important reasons: (1) to increase the reliability of responses; (2) to allow for quantitative analyses of the data; and (3) to reduce the influence of the retrospective memory component of the task. In both tests, the prospective memory retrieval cue ("res") was deliberately selected to be as simple and as easy to remember as possible so that failures in responding to the cue could be attributed to a failure in prospective memory retrieval, not a failure in remembering what the cue was. In addition, the instructions for both prospective memory tasks were detailed and administered thoroughly (see procedure and Appendix D for details), a number of steps were taken in carefully recording responses, measuring performance, 118 and calculating test scores (see results section for details on measuring performance), and a post-study recall test was included (see results and Appendix D) in an attempt to make sure prospective memory retrieval was measured as directly as possible. The Prospective Memory Questionnaire (PMQ; Hannon et al., 1995) (see Appendix D) was administered to obtain an index of self-perceptions of prospective memory task competency. This questionnaire was included in order to determine how meta-cognitive evaluations of prospective memory functioning relate to objective assessments of prospective memory retrieval. The P M Q is a 52-item questionnaire designed to assess self-rated memory abilities, and takes approximately 8-10 minutes to complete. The items pertain to memory-related behaviours and are rated on a 9-point Likert-type scale (with an anchor point of 0). Specifically, it measures self-ratings of everyday prospective memory functioning and includes items tapping into four factors, including long-term episodic, short-term habitual, and internally cued prospective memory task performance as well as techniques to remember future intentions. The internal consistency and test-retest reliability of the P M Q have been found to be high (.92 and .88, respectively; Hannon et al., 1995). Individual difference measures. A modified version of the Visual Search and Attention Test (VSAT; Trenerry, Crosson, DeBoe, & Leber, 1990) and the Dissociative Experiences Scale (DES; Bernstein & Putnam, 1986; Carlson & Putnam, 1993) were included to provided indices of ability and tendency to notice stimuli embedded within one's context/ environment. The North American Adult Reading Test (NAART; Blair & Spreen, 1989; Spreen & Strauss, 1998) was administered to provide a measure of verbal ability and premorbid intelligence, and the White Bear Suppression Inventory (WBSI; Wegner & Zanakos, 1994) was used to measure participants' general tendency 119 to suppress thoughts. See Appendix D for copies of these tests, and see Chapter 7 for details on the rationale for including each in this study. The V S A T (Trenerry et al., 1990) is generally used to assess visual scanning and sustained attention in adults. A slightly modified version of this test was constructed for the present study. This modified VSAT consisted of four different visual cancellation tasks that required respondents to cross out letters and symbols identical to a specific target. Each task consisted of 10 rows, each containing 40 stimuli. There were 10 randomly placed targets in each row. Participants were given 60 seconds to complete each of the four tasks. Tasks 2 and 3 differed from 1 and 4 in that they were accompanied by instructions to divide attention. Specifically, for these tasks, participants were asked to count backwards out loud beginning at 75 and 94 (for tasks 2 and 3 respectively) while crossing out the specified targets. Thus, the visual search test included measures of both selective and divided attention. The versions of the test administered to the younger and older adults were identical except for the font size. In the younger adult version, the font size of the stimuli was 12 pts., whereas in the older adult version, the font size was 13.5 pts. The purpose of increasing the font size for the older adults was to reduce the likelihood that problems with eyesight commonly experienced by the elderly (e.g., presbyopia) would interfere with task performance. The entire test takes approximately 8-10 minutes to administer and is scored by calculating the total number of targets correctly crossed out and subtracting the number of non-targets crossed out and the number of targets missed within the allotted time. Separate scores for the selective and divided attention tasks also are calculated in this manner. Test-retest reliability of the original V S A T is .95 (Trenerry et al., 1990). 120 The D E S (Bernstein & Putnam, 1986; Carlson & Putnam, 1993) is a 28-item, self-administered questionnaire measuring tendency toward dissociation (separation of attention, thoughts, emotions, and experiences from consciousness and/or memory). It also taps into susceptibility to suggestion and distractibility. Some items relate to common dissociative experiences (e.g., driving and not recalling part of the trip) whereas others relate to more severe dissociative experiences (e.g., out-of-body experiences). Respondents simply indicate the percentage (0% -100%) of the time they have such experiences and an overall mean score is calculated. This mean score is associated with degree of dissociation. Its reliability, internal consistency, and construct validity have been demonstrated (e.g., Carlson & Putnam, 1993). In clinical contexts, a DES score of 30 or greater suggests the presence of a dissociative disorder (Bernstein & Putnam, 1986; Foa & Rothbaum, 1998). The NAART (Blair & Spreen, 1989; Spreen & Strauss, 1998) provides an estimate of premorbid intellectual ability. It requires pronunciation of 61 different irregular rare English words (e.g., gaoled, epitome, caveat) and takes approximately 8-10 minutes to administer. Accuracy of pronunciation is used to predict IQ, which is possible given the high correlation between reading ability and intelligence in the normal population (e.g., Crawford, Stewart, Cochrane, Parker, & Beeson, 1989). The test is scored by tabulating the total number of errors (i.e., incorrectly pronounced words). The NAART correlates highly with measures of intelligence and its scores can be converted into estimated WAIS-R verbal, performance, and full-scale IQ scores. Its reliability, validity and internal consistency are high (Spreen & Strauss, 1998). The WBSI (Wegner & Zanakos, 1994) is a 15-item questionnaire measuring people's general tendency to suppress thoughts. It asks individuals to indicate on a 5-point scale the extent to which they agree (1 = strongly agree; 5 = strongly disagree) 121 with statements such as There are things I prefer not to think about', 'I have thoughts I cannot stop', and 'There are thoughts that keep jumping into my head'. Responses are summed to yield a total score that ranges from 15 to 75. The measure shows good internal consistency and test-retest reliability as well as construct validity, correlating meaningfully with measures of obsesssive-compulsiveness, anxiety, emotional vulnerability, and intrusive thinking (Muris, Merckelbach, & Horselenberg, 1996; Wegner & Zanakos, 1994). Interestingly, scores on the WBSI have been found to be positively correlated with scores on the DES (van den Hout, Merckelbach, & Pool, 1996). Procedure Participants were tested individually and the administration of the full set of tests took approximately 1 to 1 1/2 hours to complete. As can been seen in Table 4, the basic order in which the tests were administered was as follows: 3MS (as required), rating and counting task (study), prospective memory task 1 instructions, demographics questionnaire, WBSI, word completion, cued recall, word recognition, prospective memory task 2 (reading task) instructions, modified V S A T , prospective memory task 2, NAART, PMQ, and DES. Participants were informed that the study involved performing a number of cognitive tasks. After they provided their informed consent (see Appendix C for the student version of informed consent), the participants were given general instructions concerning the basic format and order of the tests. No mention of a memory test was made at this time in order to maintain the integrity and to properly administer the implicit and prospective memory tests, although memory tests were mentioned briefly in the informed consent. The first task administered was the rating and counting task. Participants were given two practice trials, one for each 122 study orientation. The words were presented individually and participants were instructed to read each word aloud, rating half according to level of pleasantness and counting the number of consonants in the other half (presented in random order). Participants wrote their responses in the test booklet provided. Once the study phase was complete, the instructions for the first prospective memory test were then given. Participants were instructed that while working on all the experimental tasks during the rest of the testing session, every time they see the three letters "res" at the beginning of a word or a word puzzle, they should spell aloud the word associated with those three letters. To ensure the participants encoded and comprehended the task requirements, they were asked to repeat the instructions to the experimenter. Specifically, participants were instructed as follows: For the next several tasks, most of the time I will give you instructions on what I want you to do immediately before each task is to be completed. However, some of the time you will be expected to remember things on your own. For example, a little later there will be a number of different tasks in which you will be shown some word puzzles. Some of them will be word fragments with missing letters like this one (note: the participant is shown the example doc_o_). Your job will be to think of a word that successfully completes the fragment, writing down the word in your test booklet. Can you think of a word that fits this fragment? The other word puzzle will be three letter combinations or word beginnings like this one (note: the participant is shown the example bla ). Your job will be to think of a word beginning with each set of three letters and to write each word in your test booklet. Can you think of a word that starts with the three letters "bla"? You will also be asked to do other things with these word puzzles, but I will not tell you about that right now. For now, I want you to know that among the letter combinations I will show you will be the letters "res." Every time you see the three letters "res" at the beginning of one of these word puzzles, I want you to complete the puzzle according to the specific task instructions AND spell out the word aloud. Spell out the word only when it starts with the letters "res." If you can't think of a word beginning with "res", then simply spell out the letters "r-e-s." OK? (note: participant is asked if instructions are clear up to this point). In addition, a little later you will also be shown a sheet of paper with several columns of different words on it. It will look something like this (note: participant is briefly shown a word list similar to the one that will be used in the word recognition test). You will be asked to do an activity with the words listed on a sheet like this, but I will not tell you about that yet. For the present purposes I want you to know that some of the words listed in the columns will begin with the letters "res." Again, every time you see a word beginning with the letters "res", I want you to spell the word out loud. Do not simply say the word aloud, instead, in every case spell it out by saying each letter of the word aloud. Now can you tell me in your own words what are you to do when you see the letters "res" either at the beginning of a word within a list or at 124 the beginning of a word puzzle? (note: instructions are repeated if necessary). At this point, if the participant could not repeat the basic task requirements accurately or appeared not to understand the instructions, then the instructions were explained a second time. Every effort was made to ensure that the participant comprehended the task directions fully. Once it was clear the participant understood the instructions, they were summarized one last time as follows: OK. To summarize, from now on, every time you see the letters "res" at the beginning of a word or at the beginning of a word puzzle, you know you have to make a response. It is up to you to remember these instructions. I will not remind you about this or repeat the instructions. Every effort was made to ensure that the instructions were completely understood. This was necessary to reduce the possibility that failures in responding to the prospective memory cue were due to a failure in retrospective memory (i.e., not remembering the content of the intention). After the instructions for the first prospective memory task were given and participants understood what was expected of them, they were given the demographics questionnaire and the WBSI to fill out, which, combined, took approximately 10 minutes to complete. The experimenter read the questions along with the participant to ensure all questions were answered. After the WSBI was filled out, participants began the word completion test (test of implicit memory). They were shown a total of 30 word fragments and three-letter word stems presented one at a time with each fragment and stem printed individually in the center of a page within a binder. The order of the presentation of the stems was randomized, except for the placement of the target stems and fillers. The participants were asked to write down, as quickly as possible, the first word that popped to mind 125 that created a word with the letters specified in each fragment and word stem. They were told that any English word was acceptable so long as it was not a proper noun. They also were asked not to repeat any of their answers. Before beginning the task, participants were given an opportunity to practice with one word fragment and one word stem (neither was included in any of the word lists). Immediately after they finished completing the stems for the implicit memory test, they were given the instructions for the cued recall task (test of explicit memory). Participants were informed that the purpose of this task was to test their memory. They were instructed to think back to the words presented on the cue cards in the rating & counting task and, using the word fragments and stems as clues, to try to remember as many words as they could from that task. They also were instructed that not all of the items related to the previously presented words and, as a result, to complete the fragments and stems only with words that they actually remember as being in the previous list. Furthermore, they were asked not to make any guesses. Again, the word fragments and three-letter stem cues were presented one at a time centered on separate pieces of paper organized in a binder. Recall of as many words as possible was encouraged by not proceeding on to the next task until the participant did not write any words for a full minute or indicated that he/she could not remember any additional words. Once the cued recall task was finished, the word recognition test was introduced. One of the two recognition test forms was presented from the binder (depending on which word list was studied) and participants were instructed to write down in the response booklet any of the words they recognized from the rating and counting task. Specifically, they were instructed to look at each word carefully and decide if it was presented in the original study list, trying not to make any guesses. All 126 words were numbered and participants were asked to write all the words they recognized in the response booklet next to their corresponding numbers. For the words that were not recognized, the response space was simply left blank. Again, the experimenter let the participants work at this task until no response had been made for one full minute or until they stated that they were finished. Following the recognition task, participants were given the instructions for the second prospective memory task-the reading task. They were asked to read aloud the short story "Agnes" and were informed that the test measured verbal fluency as well as text comprehension abilities. They were instructed that, while reading the story, every time they encountered a word or word puzzle starting with the letters "res" they were to spell out the word instead of simply reading it, and then continue with the story trying not to interrupt the flow. With their consent, reading performance was audiotaped for scoring purposes. Consent for audiotaping was requested in the informed consent sheet and participants were reminded about it before the reading task began. At this time, only the instructions pertaining to the prospective memory component of the task were given. More thorough instructions specific to reading the story were given just prior to the test. The instructions for this prospective memory task were as follows: In a few minutes I'm going to get you to do a task called the "Reading Task." For this task, I will ask you to read aloud a short story called "Agnes." While reading this story, every time you encounter a word starting with the letters "res" throughout the story, instead of simply reading the word, I would like you to spell it out. For example, if while reading the story you came across the word "resume," which begins with the letters "res," you would not pronounce the word but would 127 spell it out "r-e-s-u-m-e" instead. For your information, there may be more than one word beginning with the letters "res" in the story, or there may not be. Now, can you tell me in your own words what you have to do for the Reading Task? If the participant had difficulty repeating the task requirements or appeared not to understand what was expected for the task, the instructions were repeated until the participant could state them correctly. Again, in order to minimize the possibility of prospective memory errors occurring due to a failure in retrospective memory (i.e., not remembering what was supposed to be done), every effort was made to ensure that the instructions were completely understood. After the instructions for the reading task were given, and before the reading of the story began, participants were asked to complete the modified VSAT , which took approximately 10 minutes to complete. After that task was concluded, the participants were handed a copy of the story Agnes and given further instructions: This reading task is designed to test verbal fluency and text comprehension. Therefore, I'm going to ask you to read aloud the short story in front of you. Try to read it with average speed, neither very quickly nor very slowly. While reading you may, of course, make some mistakes but don't get too anxious or worried about them, just continue on with your reading. In short, just read this story as you would perhaps read a fairy tale to a child at home. I would also like you to keep in mind that your comprehension of the study will be tested at the end of the task, so you will also have to pay attention to 128 the content of the story. With your permission, I will audiotape your reading so it can be scored more accurately at a later time. When the participant was ready, the experimenter turned on the audio recorder and asked him/her to begin reading aloud. The story took approximately six minutes to narrate. After the reading was finished, participants were given a short comprehension test comprised of three questions testing understanding of the plot and memory for the content of the story. The questions and scoring guidelines are included in Appendix D. In addition to the comprehension test, participants were asked several general questions concerning subjective task difficulty, interest, absorption and willingness to perform the task, were given a brief test of memory for the prospective cue (i.e., the retrospective component), and asked several questions concerning subjective accuracy of performance on the task. These questions are also included in Appendix D. Once the reading task was finished, the NAART, the P M Q , and the DES were administered (in that order). After all the tasks were completed, the participants were given a debriefing sheet (see Appendix F for the student version), which was read aloud by the experimenter. The experimenter also elaborated on the background, purposes, and predicted results of the study and addressed any questions or concerns about the experiment. The debriefing sheet included three references to articles about prospective memory and implicit memory in case individuals were interested in following up on these topics of research. Participants were given a copy of the debriefing sheet to keep and thanked for their participation. 129 Results Given that the manner in which a test is scored can affect the pattern of results obtained, a number of techniques for scoring performance on the memory tests were employed. Applying a variety of scoring methods and looking at the data from different perspectives was especially important for a rich understanding of the patterns of performance on the prospective memory tests. As stated earlier, because research on prospective memory is relatively new (and challenging), experimental paradigms and research methodologies are still being explored and developed. In fact, one of the greatest barriers to conducting research on prospective memory has been the difficulty in devising precise ways to test and measure prospective memory retrieval (i.e., the prospective memory component) directly, without the potentially confounding influence of the retrospective component of the task. Therefore, the first several sections of the results section are devoted primarily to outlining in detail the methods used to score the various memory tests and presenting initial analyses of memory performance, with particular detail given to the prospective memory tests. Following these preliminary analyses, the main results-those concerning the tests for functional and stochastic dissociations-are presented, and then the results concerning self-perceptions and individual differences in prospective memory performance are reported. At the end of the results section the research findings are interpreted and discussed. Unless otherwise noted, the alpha level for all statistical analyses was set at .05. Scoring and Initial Analyses of Retrospective Memory Performance Word completion. To score performance on the implicit word completion test, a total of three scores were calculated initially for each participant: the number of nonstudied target words produced by chance (baseline) and the number of studied 130 target items correctly completed for each of the study orienting conditions. These three scores were then transformed into proportion correct and are displayed in the top part of Table 5. From the target and baseline scores, two sets of corrected scores for each participant were calculated. For one set, each individual participant's performance was used as the correction factor. The proportion of baseline items that was correctly produced by a subject was subtracted from that subject's two target proportion correct scores. This within-subject correction technique has been used in past studies (e.g., Light & Singh, 1987) and yields subject-corrected scores. Chance or baseline performance on the word completion test for calculating subject-corrected scores was .24, averaged across age groups and across the pleasantness rating and consonant counting study conditions (SD = .12; range = .00 to .62). The likelihood of producing target words in the completion task in the absence of previous study was significantly different from zero, t(131) = 22.62, p_ < .001. Baseline performance was comparable across the age groups, t(130) = -.66, p. = .508. In the case of the young participants, the completion rate for nonstudied items was .23 (SD = .10) and in the case of the older participants it was .24 (SD = .14). It has been argued that the use of subject-corrected scores as a baseline estimate for guessing may not be the best method for correcting for guessing. This is because the subject-corrected technique results in items different from the targets serving as baseline corrections for guessing within each subject. A second method for correcting for guessing is to use the items produced by subjects who received for their baseline the items that other subjects received for their targets. Although a between-subjects correction measure is used, the same items serve as both baseline and target. In the present experiment, after all counterbalances were considered, for each 131 of the young and older age groups there were groups of approximately 16 subjects who received the same baseline items that another group of approximately 16 subjects received as targets. Thus, a second set of scores was derived by subtracting the proportion of nonstudied targets produced at baseline from the proportion of studied targets correctly produced by matching subjects. These scores are referred to as item-corrected scores. The overall chance or baseline performance on the word completion test used in calculating item-corrected scores was .22 (SD = .14, range = .00 to .57), which was significantly different from zero, t(131) = 22.62, p. < 001. Baseline performance was comparable across the age groups, t(130) = -.66, p. = .508. For the young participants, the completion rate for nonstudied items was .21 (SD = .13) and for the older participants it was .23 (SD = .16). To summarize, although both types of correction scores involve subtracting unstudied baseline items from the number of targets correctly completed, in one case (item-corrected) a between-subjects measure is used where the baseline items are identical to the targets, and in the other case (subject-corrected), a within-subjects measure is used where the baseline items are different from the targets. All relevant statistical analyses on word completion performance were carried out using both the item-corrected and subject-corrected scores. As can be seen in Table 5, the two scores were very similar, and patterns of statistical findings did not differ according to which score was used. Therefore, because they are based on the probability of producing specific targets, item-corrected scores were used in all subsequent analyses. Cued recall. For the cued recall test, a total of two scores were calculated initially for each participant: the number of studied target words correctly recalled for 132 both the semantic (pleasantness rating) and nonsemantic (consonant counting) study conditions. These scores were transformed into proportion correct. The overall proportion of correctly recalled target words was calculated as well. In order to take guessing into account, a corrected cued recall score collapsed across study condition also was calculated. This corrected score reflects the number of correctly recalled studied words minus the number of incorrectly recalled nonstudied words (guesses) converted into a proportion correct. With respect to guessing, younger (M = 2.00; SD = 2.64) and older (M = 1.53; SD = 2.37) participants did not differ in the number of incorrect (nonstudied) responses produced on the task, F(1,130) = 1.16, p_ = .284. Table 5 shows the breakdown of the scores on the cued recall task. Word recognition. As with the cued recall test, separate scores on the word recognition test were calculated for the two study orientation conditions. For each study condition, the proportion of correctly recognized previously studied words was calculated. In order to correct for guessing, an overall corrected word recognition score was calculated collapsed across study condition. This corrected score reflects the number of correctly recognized studied words minus the number of incorrectly recognized nonstudied words (guesses) converted into a proportion correct. In contrast to the cued recall task, younger and older participants did differ in their level of guessing. Older participants (M = 2.52; SD = 2.52) falsely recognized a greater number of nonstudied words than did younger participants (M = 1.39; SD = 2.14), F(1,130) = 7.59, 2 = .007. An overview of the various scores for the word recognition task is given in Table 5. 133 Scoring and Initial Analyses of Prospective Memory Performance: Test 1 Prospective memory test scores. Each participant received a total of nine prospective memory targets during the first prospective memory test: three in the word completion test, three in the cued recall test, and three in the word recognition test. Two measures of prospective memory test performance were tabulated. The first measure was simply the proportion of times participants remembered to spell a word aloud when the target cue ("res") occurred (i.e., strict scoring or carrying out the task exactly as instructed). The second measure involved a somewhat more detailed scoring scheme (i.e., lenient/ detailed scoring). Each cue presentation was scored as follows: 3 points were assigned if the participant remembered to spell the appropriate word out loud in response to the presentation of the target cue ("res"); 2 points were assigned if the participant responded to the target cue in way that was different from what was instructed (e.g., simply saying the word aloud, or indicating that something should be done); 1 point was granted if the participant responded to the wrong cue (e.g., "rel") or responded late (after the participant had moved onto subsequent items); and a score of zero (0 points) was given if the participant failed to respond at all. The total score was expressed as the proportion of points awarded out of a maximum of 27 points. A total of 103 (78%) participants made at least one response in the first prospective memory task. A total of 29 (22%; 9 young, 20 old) participants failed to make any responses and 3 (2.3%; 2 young, 1 old) responded to all nine cues. Overall, across participants and experimental conditions, mean performance according to the strict scoring criterion was .30 (SD = .28), which was significantly lower than mean performance as scored by the lenient/ detailed method (M = .35, SD = .27), t( 131) = 134 6.68, p_ < .001. The top of Table 6 gives a summary of both methods of scoring the first prospective memory task broken down according to age group. Again, for each age group, the strict scoring scheme resulted in significantly lower mean scores than the lenient scoring scheme, t(65) = 3.72, p_ < 0001 and t(65) = 5.80, p. < .001 for the younger and older groups, respectively. These findings are consistent with Dobbs and Rule (1987) who found prospective memory task scores to be higher when scored according to lenient scoring criteria as opposed to strict criteria. Unless otherwise stated, statistical analyses were conducted using scores calculated using the lenient criteria given that they provide a more detailed and somewhat more sensitive measure of memory performance and the patterns of results were very similar when strict scores were used. Arguably, one of the major issues with constructing tests to measure prospective memory functioning has centered around assessing whether the to-be-performed intention leaves the participant's conscious awareness during the retention interval and performance of the ongoing or "cover" task. In scoring performance on the prospective memory tasks, all instances in which participants mentioned intention-relevant information when a prospective memory cue was not present were recorded (e.g., "Oh, this is the task that I have to spell out 'res' words," "Is this the Yes' task?" "OK, I better not forget to do the 'res' thing," "Darn, I'm supposed to be spelling words now," etc.). A total of 42 participants (32%) made some reference to the to-be-performed intention at least once between the time the prospective memory task instructions were administered and the end of the prospective memory task. Although this is a rough index of whether the intention popped to mind in the absence of the retrieval cue, it does suggest that more than two-thirds of participants may not have had (or perhaps had fewer) intention-related thoughts in the absence of target cues. 135 To test whether such rehearsals or reminders were associated with greater success on the prospective memory test, a point-biserial correlation between whether participants did or did not make mention of the intended action and their score on the actual test. The analysis revealed a significant positive relationship, r p b(132) = .287, p < .001. Interestingly, although a greater number of younger participants (n = 25) made reference to the intention than did older participants (n = 17), the correlation was significant only for the older group, rp b(66) = .161, p. = .196 and r p b(66) = .378, p = .002 for the young and older groups, respectively. Scoring and Initial Analyses of Prospective Memory Performance: Test 2, The Reading Task Prospective memory test scores. The prospective memory target cue (words beginning with the letters "res") occurred a total of nine times in the reading task. Similar to the first prospective memory test, several measures of performance were calculated. The first measure was simply the overall proportion of times the participant responded correctly to the prospective memory cue (i.e., strict scoring). The second measure utilized the same scoring scheme that was applied to the first prospective task (i.e., lenient/ detailed scoring). Specifically, for each target cue presentation, 3 points were assigned if the participant remembered to spell the appropriate word out loud in response to the presentation of the target cue ("res"). Two points were assigned if the participant responded to the target cue in a way that was different from what was instructed (e.g., repeating the word, or indicating that something should be done). One point was granted if the participant responded to the wrong cue (e.g., a word beginning with "rel") or responded late (after the participant had moved onto subsequent items), and a score of zero (0 points) was given if the participant failed to 136 respond at all. The total score was expressed as the proportion of points awarded out of a maximum of 27 points. A total of 94 (71.2%) participants made at least one response in the second prospective memory task. A total of 38 (28.8%; 8 young, 30 old) participants failed to make any responses and 29 (21.9%; 19 young, 10 old) responded to all nine cues. Overall, across participants and experimental conditions, mean performance according to the strict scoring criterion was .46 (SD = .37), which was significantly lower than when performance was scored with lenient/ detailed criteria (M = .51, SD = .39), t(131) = 8.54, p < .001. Table 6 provides a summary of the scores for the second prospective memory task broken down according to age group. Strict scores were significantly lower than lenient scores for both age groups, t(65) = 6.47, p < 0001 and t(65) = 5.62, p < .001 for the younger and older participants, respectively. Again, because scores based on lenient scoring criteria are more sensitive and detailed indices of prospective memory test performance, they will be used for the majority of analyses. Reading time and self-report measures. Since reading the story Agnes in the second prospective memory task was a self-paced task, possible differences in reading time (expressed in seconds) between the younger and older participants were explored. Overall, across all participants, the story took an average of 365.70 seconds (SD = 65.55) to read with a range of 254 to 645 seconds. On average, older participants (M = 386.27; SD = 69.23) took longer to read the story than did younger participants (M = 343.12; SD = 54.04), F(1,130) = 15.93, p_ < .001. Nevertheless, follow-up questions showed that the two age groups did not differ in self-rated difficulty of simultaneously reading the story and spelling aloud all words beginning with "res", F(1,89) = 1.16, p_ = .285, and in self-rated willingness to comply with the task of spelling 137 out all words beginning with the letters "res", F(1,89) = 3.90, p = .061 (see Table 7). However, older adults did rate themselves as being somewhat more interested and absorbed in reading the story than did younger adults, F(1,130) = 14.88, p_< .001 and F(1,130) = 13.38, p < .001, for the interest and absorption ratings, respectively. This suggests the possibility that compared to the younger adults, the story may have been more of a distraction for the older adults, decreasing the likelihood of noticing prospective memory cues. Yet, examination of the story comprehension scores (maximum score = 6; see Appendix D for scoring system), which can be considered as a second, and more objective measure of absorption, indicated that older participants scored significantly lower than younger participants, F(1,130) = 20.36, p < .001 (see Table 7). As mentioned previously, one issue concerning the measurement of prospective memory functioning is the extent to which the intended action enters/ dominates conscious awareness during the retention interval and performance of the ongoing or "cover" task. Interestingly, out of the 91 participants who made at least one response, only 23 participants (25%) reported that they were constantly thinking about their intention to spell out words beginning with "res" while reading the story, whereas 68 participants (75%) reported remembering the intention only when they encountered the target cue. In addition, the number of younger (n = 13 or 23%) and older (n = 10 or 29%) participants who reported constantly thinking of the intention during the reading task was quite comparable, as was the number in each group who reported remembering the intention only when they noticed the cue (n = 44 or 77% for the younger group and n = 24 or 71 % for the older group). In order to determine whether reports of constantly thinking about the intention were related to success in the task, a point-biserial correlation was computed. Surprisingly, there was no relationship, rp b(91) 138 = .063, p_ = .551. In addition, as with the first prospective memory task, all instances of intention-related comments made from the time the task instructions were given until the end of the task were recorded (e.g., "oh yeah, this is the 'res' task"--stated in the absence of target words). However, unlike the first prospective memory task but consistent with the findings concerning whether subjects constantly thought about the intention, intention-related comments were not associated with higher task scores, r p b(132) = .02, p = .824. This lack of association held true for both the younger and older age groups, r p b(66) = .095, p_ = .449 and r p b(66) = .196, E = -115, respectively. Follow-up questions for the reading task also included several questions directed at participants' self-assessment or evaluation of their own performance in the task. Specifically, participants were asked to rate on a four-point scale (1 = never; 4 = every time) how well they did in responding to the cue "res" while reading the story. They also were asked to rate how confident they were in their answer to that question (1 = not confident; 4 = very confident). Although confidence ratings were negatively correlated with self-ratings of task performance (r[66] = -.218, p_ = .079 for young and r[66] = -.403, p_ < .001 for old), both younger and older adults' performance ratings correlated highly with their actual performance on the reading task: r(66) = .85 for the young group and r(66) = .86 for the old group, p_s < .001. In fact, when asked how many times they responded to words beginning with "res" in total, across all tasks they had completed to that point (i.e., both prospective memory tasks), participants' performance estimates were very accurate, r(130) = .80, p < .001 (r[66] = .70 for young and r[64] = .84 for the old, ps < .001). Prospective memory errors. Following the scoring system of Kvavilashvili (1998), performance on the second prospective memory task also was scored in terms 139 of types of errors. Serious errors or omissions occur when subjects fail to respond to the cue. Instead of spelling out the words beginning with the letters "res", they simply read the words and continue reading without remembering that they were supposed to respond to the cue. Late responses include errors of two types: intermediate errors and slight errors. Intermediate errors occur when subjects initially read a word beginning with "res" but then become aware of their mistake and quickly make the correct response. Slight errors occur when subjects realize they are making a mistake in the very course of reading the target word and immediately correct it. Precise scoring of these types of errors was ensured by having: (1) the experimenter record responses in detail at the time of testing, and (2) a second coder double-check the accuracy of the originally recorded responses by listening to the audiotapes of participants' task performance. Out of the 132 participants who completed the task, 43 (32.6%) forgot to spell out target words on all 9 occasions, whereas 11 (8.3%) remembered on every trial. Performance differed considerably for the younger and older age groups. Within the young group, only 9 (13.6%) participants forgot to respond to target cue words every time, 7 (10.6%) remembered to respond on every trial, 33 (50%) forgot their intention on one to three trials, and 17 (25.8%) forgot on four to eight occasions. By contrast, within the older group, 34 (51.5%) participants failed to respond to any of the prospective target words, 4 (6.1%) responded to all the targets, 15 (22.7%) forgot their intention on one to three trials, and 13 (19.7%) forgot on four to eight trials. The bottom of Table 6 shows the mean number of the different types of errors committed by participants according to age group. In order to see if the difference in the number of errors made by younger versus older participants was statistically significant, the total number of prospective memory errors in the reading task was 140 entered into a one-way A N O V A with age group (young and old) as the independent variable. The analysis revealed a significant difference, with younger participants making almost half the number of errors as older participants, F(1,130) = 24.49, 2 < .001. In order to examine types of errors in more detail, analyses were conducted separately on serious errors/ omissions, slight errors, and intermediate errors. The older group made significantly more serious errors than the younger group, F(1,130) = 22.32, 2 < .001, and although the groups did not differ in the number of intermediate errors they made, F(1,130) = .96, 2 = -33, the younger age group made significantly more slight errors than the older group, F(1,130) = 4.95, 2 < -05 (see Table 6). This suggests that, compared to older participants, younger participants were more likely to "catch" themselves and notice target words in the course of reading them as signs of the previously formed intention. Source of prospective memory errors. As discussed in Chapter 2, one challenge in testing prospective remembering is teasing apart the retrospective and prospective components of the prospective memory task. Many researchers have pointed out the problem of identifying the source of prospective memory errors. In other words, if a participant forgets to carry out an intention, is it possible to determine whether the memory failure was due to forgetting that something was to be done (i.e., the intention) or to forgetting exactly what it was that had to be done (i.e., the content of the intention). In an attempt to address this issue, all participants were asked a series of follow-up questions designed to assess their memory for the retrospective component of the task (i.e., what they were supposed to do). When asked if they were instructed to do something extra while reading the story, 61 (92.4%) young participants and 53 (80.3%) older participants answered "yes." Of those who were unsure or did 141 not remember (5 young, 13 old), upon further questioning and cueing all 5 of the younger adults and 11 of the older adults remembered correctly that the instructions were to spell out words beginning with the letters "res". Only two of the older adults did not remember exactly what it was they were supposed to do. This provides evidence that practically all of the failures to respond to the target cues in the reading task were failures of prospective, and not retrospective, memory retrieval. Initial Performance. Forgetting, and Recovery Scores for Prospective Memory Tests 1 and 2 Given that both prospective memory tests were repeated-intention designs, initial performance scores, subsequent forgetting scores, and recovery scores were calculated (see Maylor, 1993a) for each test. Initial performance scores indicate whether participants responded to the very first cue presentation, and are important to examine because they are uninfluenced by previous responses. The percentages of participants who responded to the first prospective memory cue (as measured by lenient scoring criteria) were categorized according to age group for both prospective memory tasks and are shown in Table 6. For both tasks, the younger participants were far more likely to notice and respond to the very first prospective memory cue presented than were the older participants. Response frequencies were comparable across prospective memory tasks for both the younger and older groups. If a participant succeeded in responding to the target cue at one time, but then failed to respond to it the next time it was presented, this was considered an instance of subsequent forgetting. The probability of subsequent forgetting was then calculated by dividing the number of instances of forgetting by the number of opportunities for forgetting. Similarly, if a participant failed to respond to one presentation of the target 142 cue, but then succeeded in responding to the next presentation, this was scored as an instance of recovery. The probability of recovery was then calculated by dividing the number of instances of recovery by the number of opportunities for recovery. (Participants with zero opportunities for forgetting or zero opportunities for recovery could not be included in the analyses of forgetting and recovery, respectively). The mean probabilities of forgetting and recovery for the young and old age groups are shown in Table 6. Separate ANOVAs performed on these scores revealed no significant differences between the age groups for either prospective memory test. For prospective memory task 1, subsequent forgetting scores did not vary according to age, F(1,97) = .004, p. = .950, nor did recovery scores, F(1,71) = 1.17, p. = .283. Similarly, for prospective memory task 2, neither subsequent forgetting scores, F(1,59) = .475, p = .493, nor recovery scores, F(1,55) = .355, p. = .554 showed an age effect. Thus, younger and older adults did not differ in their probabilities of forgetting and recovering their intentions in both of the prospective memory tasks. It is noteworthy that the probability of forgetting was somewhat higher in the second prospective memory task (reading task) than in the first. Functional Dissociations?: The Effects of Age. To evaluate differences between young and older participants on the various memory tests, a one-way multivariate analysis of variance (MANOVA) was performed. Age (young and old) was the between-subjects independent variable and scores on word completion, cued recall, word recognition, prospective memory test 1, and prospective memory test 2 (reading task) were the dependent variables. All memory test scores are expressed in terms of proportion correct. Item-corrected scores for the word completion test, corrected (hits - false alarms) scores for cued recall and word 143 recognition tests, and lenient scores for the prospective memory tests were utilized for these analyses. The same pattern of results was found when a second MANOVA incorporating the alternate scores for word completion (subject-corrected), cued recall (proportion targets produced), word recognition (proportion targets produced), and prospective memory tests 1 and 2 (strict scoring criterion) was conducted. For ease of exposition, only the former findings are reported inasmuch as identical results were obtained for both analyses. There were no univariate or multivariate within-cell outliers at p_ = .001. Results of the evaluation of assumptions of normality, homogeneity of variance-covariance matrices, linearity, and multicollinearity were satisfactory. With the use of Wilks' criterion, the combined dependent variables were significantly affected by age, Wilks' X = .606, F(5,126) = 16.41, p < .001. Table 8 displays the mean scores and standard deviations for the dependent variables. As the table indicates, according to follow-up univariate analyses, age had a significant effect on all five memory tests: word completion, £(1,130) = 9.03, p < .001; cued recall, F(1,130)= 16.50, p < .001; word recognition, F(1,130) = 65.39, p < .001; prospective memory test 1, F(1,130) = 13.48, p_ < .001; and prospective memory test 2, F(1,130) = 24.40, p < .001. Younger participants scored significantly higher than older participants in all of the memory tests, indicating an absence of any dissociations between the prospective and retrospective memory tasks due to age. Hedge's d was used to calculate effect sizes for the age differences in performance on each of the memory tests (see Table 8). Word recognition showed the largest effect size, followed by prospective memory task 2, cued recall, prospective memory task 1, and word completion. 144 Although significant differences between younger and older adults on the implicit memory test and even on the word recognition test (e.g., Craik, 1986; Craik & McDowd, 1987), were not expected, they are not unusual (e.g., Chiarello & Hoyer, 1988; La Voie & Light, 1994; Light & Singh, 1987). However, as discussed previously, it is more often the case that the performance by young and old on these types of memory tests is quite comparable (see Craik, 1986; Graf, 1990; Light, 1991; Salthouse, 1982). In order to get a more detailed picture of the effects of age on the various memory tests administered, a third age group was created that included the oldest participants in the sample. Previous research has shown that substantial and even quite dramatic declines in memory performance can occur in later years, especially over the age of 80 (e.g., Erngrund, Mantyla, & Nilsson, 1996; Mantyla & Nilsson, 1997; Uttl et al., 1999). Therefore, the old age group was re-categorized according to a median split (median age = 76), resulting in a young-old group with a mean age of 67.7 years (SD = 4.12, n = 34) and an old-old group with a mean age of 81.5 years (SD = 4.50, n = 32). By categorizing participants in this manner, it was possible to separate the effects of the "oldest" old on the older age group and reassess patterns of age differences. However, it must be noted that dividing the older age group in this way reduces the size of the comparison groups and reduces the power of the statistical analyses to find group differences. Similar to the previous analysis, a MANOVA was conducted with age (young, old, and oldest) as the between-subjects independent variable and scores (proportions correct) on word completion, cued recall, word recognition, prospective memory test 1, and prospective memory test 2 (reading task) as the dependent variables. Results of the evaluation of assumptions of normality, homogeneity of variance-covariance matrices, linearity, and multicollinearity were satisfactory. Using Wilks' criterion, the 145 overall MANOVA was statistically significant, indicating an effect of age on the combined dependent variables, Wilks' X = .575, F(10, 250) = 7.97, p < .001. As can be seen in Table 9, follow-up univariate analyses revealed significant age effects on all of the dependent variables: word completion, F(2,129) = 4.49, p < .01; cued recall, F(2,129) = 9.83, p < .001; word recognition, F(2,129) = 32.45, p < .001; prospective memory test 1, F(2,129) = 8.54, p < .001; and prospective memory test 2, F(2,129) = 14.36, p < .001. Bonferroni post-hoc comparisons were then conducted to identify the source of the age differences within each of the memory tests. Table 9 displays the pattern of results. Of particular interest is the finding that once the scores of the oldest age group were removed, the young and old age groups no longer differed statistically in their performance on the implicit word completion test. However, there was no difference in performance between the old and oldest group either. Only the young and oldest group showed a statistically significant difference. The two explicit memory tests, cued recall and word recognition, showed similar age differences with the younger group exhibiting better memory than both the older age groups, which did not differ significantly from one another. Also of interest was that the prospective memory tests showed different patterns of age differences. Prospective memory test 2 (reading task) scores were significantly higher for the younger participants than both the old and oldest participants, and the two old age groups did not differ. However, surprisingly, there was no significant difference in scores on prospective memory task 1 between the young and old age groups, and scores for the old and oldest age groups did not differ either. The only difference in performance on the first prospective memory task 146 that was statistically significant was the difference between the young and oldest age groups. Functional dissociations? The Effects of Levels of Processing As stated previously, the levels of processing study task was included primarily as a manipulation check for the implicit and explicit memory tests. However, since all participants studied one prospective memory target semantically (deeply) and one nonsemantically (shallowly), exploratory analyses of the effect of depth of encoding on prospective memory performance were performed. The number of times participants noticed previously studied words as prospective memory cues and responded to them in some manner indicative of prospective remembering was calculated for both the semantic and nonsemantic study orientations. These sums were then expressed as proportions by dividing by the total number of possible opportunities available to respond to each study word. As the study manipulation was applicable only to the first prospective memory test, analyses were not conducted on prospective memory test 2. The number of semantic and nonsemantic study words produced in the implicit word completion and the two explicit (cued recall and word recognition) retrospective memory tests were calculated as well (see Table 5). It is important to note that the semantic and nonsemantic prospective memory scores are contingent upon performance in the word completion and cued recall tests because in order for participants to respond specifically to those previously studied "res" words (i.e., by spelling the words aloud), they would have had to produce them for the completion and cued recall tests in the first place. However, this was not the case for the word recognition task as the nature of recognition tasks are such that the previously studied words are presented in their entirety when presented to the participants. 147 Semantic vs. nonsemantic study conditions. Using the proportion of targets produced in each of the study conditions as test scores, a 2 X 2 between-within MANOVA was performed on the four memory tests of interest: implicit word completion, explicit cued recall, explicit word recognition, and prospective memory test 1. Study orientation (semantic/ rating pleasantness, nonsemantic/ counting consonants) was the within-subjects independent variable and age (young and old) was the between-subjects independent variable. All tests of the assumptions of normality, homogeneity of variance-covariance matrices, and sphericity were found to be satisfactory. With the use of Wilks' criterion, the combined dependent variables were significantly affected by both study orientation, Wilks' k = .370, F(4,127) = 54.10, p. < .001, and age, Wilks' X = .665, £(4,127) = 16.02, p. < .001, but not by their interaction, Wilks' X = .973, F(4,127) = .87, p = .483. To investigate the impact of each main effect on the individual dependent variables, follow-up univariate tests were performed. Table 10 gives the means and standard deviations corresponding to these analyses. With respect to the main effect of age, as was demonstrated above, performance on all four memory tests differed according to age with younger participants scoring higher than older participants: word completion, F(1,130) = 9.03, p < .001, cued recall, F(1,130) = 51.41, p < .001, word recognition, F(1,130) = 38.08, p < .001, and prospective memory test 1, F(1,130) = 8.88, p < .001. With respect to levels of processing, as expected, manipulating study condition significantly affected performance on the two explicit memory tests, F(1,130) = 51.97, p < .001 and F(1,130) = 216.38, p < .001 for the cued recall and word recognition tests, respectively, with words accompanied by semantic study remembered more often than words 148 accompanied by nonsemantic study. Also, as predicted, study condition did not have a significant effect on implicit word completion test scores, F(1,130) = .12, p = .725. Interestingly, the study manipulation did not have a significant effect on prospective memory test performance either, F(1,130) = .93, p = .338, indicating that participants were not more likely to respond to prospective memory cues that had been studied previously under semantic conditions versus cues that had been studied under nonsemantic conditions. The means and standard deviations for prospective memory test 1 performance broken down according to study condition can be found in the top of Table 10. These results demonstrate dissociations between the implicit and explicit retrospective memory tests as well as dissociations between the prospective and retrospective memory tests. However, the prospective-retrospective dissociations are difficult to interpret due to the exploratory nature of the analyses and given that only one prospective memory cue word (i.e., words beginning with "res") was studied under each of the study orientations. Studied vs. nonstudied words. Since the levels of processing manipulation did not have an effect on prospective memory performance, a second analysis was conducted to test whether there was a performance advantage for previously studied target cue words (regardless of study orientation) over cue words that were not previously encountered during study. In other words, did previously studied "res" words serve as better prospective memory cues than nonstudied words beginning with the letters "res"? Thus, the number of times participants spelled out a "res" word that had been included in the previously study list (regardless of study orientation) was calculated and compared to the number of times participants spelled out a "res" word that had not been on the study list (see the bottom of Table 10 for proportions). A 2 X 149 2 between-within A N O V A was then conducted on prospective memory scores (expressed as proportion of responses made) with study condition (studied and nonstudied) as the within-subjects factor and age (young and old) as the between-subjects factor. This analysis indicated that there was no advantage in response rates for previously studied cue words, F(1,130) = .67, p = .416. However, similar to the preceding analysis, there was a main effect for age, F(1,130) = 8.96, p < .001, but no interaction between study condition and age, F(1,130) = 1.10, p = .296. Stochastic Dissociations Given that age was found to have a differential effect on memory test performance, the patterns of correlations within each of the age groups were examined. That is, the patterns of relations between scores on the various memory tests were explored within the younger and within the older age groups separately. The specific test scores used for these analyses were the lenient prospective memory scores, item-corrected word completion scores, and scores corrected for guessing for the cued recall and word recognition tests. Table 11 displays the correlation matrices for memory test scores and shows a somewhat different pattern of relations among the memory tests for the young and older adults. Interestingly, no correlation was found between performance on the two prospective memory tests for the young group, but a significant positive correlation was found for the older group (note that overall, across the age groups, r(132) = .459, p < .001). Prospective memory test performance was not related at all to performance on the implicit word completion test for either the young or older age groups. However, for the young group, performance on the first prospective memory task was positively correlated with recognition memory test scores, whereas performance on the second prospective memory task (the reading 150 task) were positively correlated with cued recall scores. The same pattern of results was found for the older group (with comparable magnitudes of association), however performance on the first prospective memory test also was significantly correlated with scores on cued recall. In order to get a clearer picture of the relationship between the different memory tests, correlations were computed between test scores while controlling for age. The bottom of Table 11 shows the pattern of partial correlations among the memory tests. When the variability due to age is controlled for, performance on the two prospective memory tests is significantly correlated. Neither prospective memory test correlated with implicit memory test scores; however, prospective memory test 1 was significantly correlated with both cued recall and word recognition and prospective memory test 2 was significantly correlated with cued recall. Although the implicit and word recognition tests did not correlate, there was a modest positive correlation between the implicit and cued recall tests. These patterns of results provide strong evidence that prospective memory test performance is related to explicit, but not implicit, memory test performance. Performance on the implicit and both the prospective tests were dissociated, whereas only the explicit word recognition showed a dissociation with prospective memory task performance (and only for prospective memory test 2). Retrieval Context and Prospective Memory Test Performance It has been proposed that one major difference between prospective and retrospective memory tests is whether the rememberer is in a "retrieval mode" at the time when memory retrieval is supposed to take place (e.g., McDaniel & Einstein, 2000a). In typical laboratory tests of retrospective memory, experimental instructions induce or put subjects into retrieval mode, which initiates remembering and sensitizes 151 subjects to the meaning of retrieved memories. By contrast, in tests of prospective memory, environmental cues must be noticed and must trigger memory for the previously formed intention, in the absence of being in retrieval mode. Nevertheless, all retrieval modes are not the same. In explicit retrospective memory tests, the retrieval mode is characterized by a conscious, deliberate attempt to recollect information from a specific prior episode. By contrast, the retrieval mode associated with implicit tests does not involve any such awareness or intentional search of prior episodes. Because in the first prospective memory test, prospective memory cues were embedded within the implicit and explicit memory tests, the opportunity is available to explore possible differences in the probability of responding to prospective cues within each of these retrieval contexts. Therefore, the proportion of correct prospective memory responses within the implicit word completion test, the explicit cued recall test, and the explicit word recognition test were calculated and compared. An overall one-way repeated measures A N O V A with retrieval context (word completion, cued recall, and recognition) as the independent variable and proportion correct (lenient scoring) on prospective memory test 1 as the dependent variable revealed a significant effect, F(2, 262) = 76.87, p. < .001. Follow-up comparisons indicated that significantly more prospective memory responses were made within the context of the implicit word completion test (M = .55, SD = .40) than the cued recall test (M = .37, SD = .38) and performance within the context of cued recall was significantly higher than when the context was word recognition (M = .14, SD = .27), p_s < .001. In order to explore the possibility of age differences in prospective memory performance within each of these "retrieval modes" a 2 X 3 between-within A N O V A was carried out on prospective memory test 1 scores with age (young and old) as the between-subjects variable and retrieval context (word completion, cued recall, and recognition) 152 as the within-subjects variable. Of course, there was a main effect for retrieval context, F(2, 260) = 77.90, p < .001, as well as a main effect for age, F(1,130) = 13.48, p. < .001. However, the interaction between context and age was not statistically significant, F(2, 260) = 2.75, p = .07. Therefore, younger and older participants showed the same pattern of responses among the different retrieval contexts. Retrieval context versus practice effects. Because the order in which the tests were administered was the same for all participants, with the word completion test occurring first, followed by the cued recall, and then the word recognition, and because prospective memory response rates decreased across these tests, it is possible that these results reflect some kind of order effect rather than an effect due to retrieval context. Although it is not possible to test this directly, the absence of practice effects (which seem more probable than a decrease in performance given the repeated-intentions design) is worth noting. The findings with respect to practice effects in prospective memory tasks have been inconsistent and appear to be primarily dependent upon the nature and characteristics of the tasks themselves. For example, whereas Maylor (1993a, 1998) reported a general increase in prospective memory performance across blocks of trials, Kvavilashvili (1998) reported a decrease in performance across trials. For comparison purposes, performance on prospective memory test 2 (the reading task) was tested for the presence of practice effects. An overall repeated measures ANOVA with trial block (first three trials, middle three trials, and last three trials) as the independent variable and proportion correct (lenient scoring) as the dependent variable indicated no significant effect, F(2, 262) = 1.39, p = .252 (M = .49, SD = .41 for the first three trials, M = .53, SD = .43 for the middle trials, and M = .49, SD = .41 for the last three trials). Analyses of variance with age group as 153 an independent variable showed a significant effect for age on performance across trials with young participants scoring higher than older participants, F(1,130) = 24.40, p. <.001, but no interaction, F(2, 260) = 1.76, p = .174. Similar results were found when the number of serious errors committed across trials was examined. Thus, for the reading task, responses to the prospective memory target cues were consistent across trials for both the young and the old age groups. Interestingly, this pattern of results is not consistent with those of Kvavilashvili (1998) who also tested prospective memory in the context of a reading task. However, her task included over twice as many prospective memory cues (20 cues in total) than the task included in this study, providing many more opportunities for forgetting (see also Ellis et al., 1999). In addition, the absence of practice effects appears inconsistent with the pattern of responding that would be predicted from the automatic associative activation model of prospective memory. According to this model, prospective memory retrieval occurs automatically. When the activation level of the cue-intention pairing in memory is high enough, the prospective target cue automatically triggers the activation of its associated intention. The activation level of the cue-intention pairing is assumed to increase with rehearsal. Thus, if the automatic associative activation model is correct, repeated-intentions designs should result in an increase in the probability of responding across cue presentations, especially for those individuals who have made at least one previous response to the prospective cue. Furthermore, a greater likelihood of ceiling effects would be expected as well. Self-Assessment of Prospective Memory Performance In addition to the objective measures of prospective memory retrieval, a measure of self-perceptions of prospective ability was included in this study. The 154 purpose of including a self-assessment of prospective remembering was to investigate whether (and to what extent) self-ratings correspond with demonstrated capabilities of prospective memory retrieval and to explore whether this relationship differs as a function of age. As outlined above, the PMQ yields a total score as well as score on the following four factors: long-term episodic, short-term habitual, internally cued performance and techniques to remember. Higher scores on the long-term episodic and short-term habitual factors indicate greater levels of forgetting. Higher scores on the internally cued and techniques to remember factors indicate greater usage of various aids to assist memory. As can be seen at the bottom of Table 6 younger participants had a higher overall mean test score as well as higher mean scores on all four of the factors than did the older participants. To test whether these differences were statistically significant, a one-way MANOVA was conducted with age (young and old) as the between-subjects independent variable and the various PMQ test scores as the dependent variables. The overall MANOVA for the combined dependent variables was statistically significant, Wilks' X = .751, £(5,126) = 8.36, p < .001. Follow-up univariate analyses indicated that younger participants' scores were significantly higher than the scores of older participants on all of the test scores: total PMQ score, F(1, 130) = 31.46, p < .001; long-term episodic, £(1,130) = 37.15, p_ < .001; short-term habitual, £(1,130) = 18.11,p < .001; internally-cued, £(1,130) = 24.03, p < .001; and techniques to remember, £(1,130) = 6.46, 2 < -01. Thus, younger participants rated themselves as being more forgetful with regard to both long-term episodic and short-term habitual prospective memory tasks than did older participants. In addition, younger participants reported using more internal cues or reminders (e.g., mental rehearsal, mental imagery) and various external cues/ techniques (e.g., reminder 155 notes, appointment books) to help them remember to carry out intentions (see Table 6 for means and Appendix Q for questionnaire items). To test the extent to which self-rated prospective memory ability was related to actual performance on prospective memory tasks, correlations were computed between scores on the PMQ and scores on prospective memory task 1 and prospective memory task 2. Correlations based on the entire sample revealed a significant positive association between total scores on the P M Q and scores on prospective memory test 2 (the reading task), r(132) = .238, p < .01, which actually can be largely attributed to the significant positive correlation with scores on the long-term episodic factor, r(132) = .259, p < .01. The other PMQ factor scores did not correlate significantly with performance on prospective memory task 2. By contrast, total scores on the PMQ were not related to scores on prospective memory test 1, r(132) = .158, p > .05, and only the techniques to remember factor correlated significantly with performance on this test, r(132) = .201, p < .05. Interestingly, no significant correlations were found between the PMQ and the prospective memory study tasks when computed for the young and older age groups separately-not even when partial correlations were calculated controlling for the variability due to age, rs ranged from .010 to . 164, ps > .05. Of course, it is important to keep in mind that the questions on the PMQ refer to everyday prospective memory tasks, which are considerably different from the prospective memory tasks carried out within a controlled laboratory setting. Individual Differences in Prospective Memory Performance To assess how younger and older participants' scores compared on the various individual difference measures administered, a one-way M A N O V A was performed. Age (young and old) was the between-subjects independent variable and scores on the 156 modified V S A T (total score plus selective and divided attention scores), the D E S , the WBSI , the NAART, and full scale, verbal, and performance IQ scores derived from scores on the NAART were the dependent variables. The combined dependent variables were significantly affected by age, Wilks' X = .354, F(6,125) = 38.02, p < .001. Follow-up univariate tests showed that age had an effect on all of the measures, except for scores on the WBSI, F(1,130) = 1.25, p = .265. As can been seen in Table 12, the younger group scored almost three times higher on the DES than the older group, F(1,130) = 51.44, p < .001. The younger participants also had higher total scores, F(1,130) = 115.84, p < .001, higher selective attention scores, F(1,130) = 127.19, p < .001, and higher divided attention scores, F(1,130) = 7.86, p < .001, on the visual search task than did the older participants. However, with respect to verbal ability and intelligence, older participants significantly outperformed the younger participants as indicated by higher NAART scores and higher estimated IQ scores, all Fs(1,130) = 14.54, p < .001. Pearson product-moment correlation coefficients were calculated for each age group to determine the relationship between the measures of prospective memory (prospective memory test 1, prospective memory test 2, self-reported prospective memory) and the individual difference variables of educational level, occupational status, self-reported health status, level of social activity, number of hours per week spent outside the home, WBSI scores, DES scores, modified V S A T scores (total score plus selective and divided attention task scores), and NAART scores. Table 13 presents the intercorrelations for the young group and Table 14 displays the intercorrelations for the older group. 157 For the young participants, prospective memory test 1 scores were positively correlated with number of hours per week spent outside the home. They were also correlated with WBSI scores suggesting that a tendency toward thought suppression is related to success in noticing and responding to target cues within the context of the first prospective memory task. Young participants' total V S A T scores and divided attention scores were positively related to performance in prospective memory test 2 (the reading task), which is not surprising given that success in the task relies on the ability to divide attention between reading the story and noticing prospective target cues. Correspondingly, the young participants' scores on the retrospective memory cued recall test were positively correlated with V S A T total and divided attention scores (see Table 15), and word recognition scores were positively related to VSAT divided attention scores as well. Higher scores on cued recall also were associated with higher verbal ability as measured by the NAART. Some interesting trends were observed with regard to scores on the PMQ. Younger participants' total P M Q score was positively correlated with general health and DES scores and negatively correlated with V S A T total and NAART scores. Higher scores on the P M Q reflect, in general, higher levels of self-reported forgetting. Higher self-rated health, a greater tendency to dissociate, poorer visual search abilities, and poorer verbal ability all were related to higher levels of self-reported prospective memory failures. Higher DES scores also were related to self-reported forgetting on long-term episodic and short-term habitual prospective memory tasks as well as a greater reliance on internal cues to remember previously formed intentions. A greater tendency toward thought suppression also was correlated with internal cue use for the younger participants. For the older subjects, success on both prospective memory task 1 and 2 (reading task) were positively associated with higher levels of education, higher VSAT 158 scores (total, selective, and divided), and higher NAART scores (verbal ability). In addition, prospective memory test 1 performance was negatively correlated with self-rated general health, and prospective memory test 2 performance was positively associated with number of hours per week spent outside the home (see Table 14). Consistent with the prospective memory scores, older adults' scores on retrospective cued recall were positively associated with all three V S A T scores (total, selective, and divided attention conditions) as well as NAART scores (see Table 16). Surprisingly, older adults' implicit memory test performance was negatively related to general health ratings and ratings of the extent to which health interferes with participating in activities. Similar to the younger adults, older adults' PMQ scores were positively related to the WBSI and the DES (all except for forgetting on short-term habitual tasks). Thus, higher tendencies toward dissociation and thought suppression were related to higher PMQ total scores, increased forgetting on long-term episodic prospective memory tasks, and an increased tendency to rely on both internal cues and external reminders. PMQ total, long-term episodic, and use of internal cues also were associated with ratings indicating better health relative to others the same age. To summarize, several clear trends with regard to the relation of individual difference factors to prospective memory performance can be noted. First, visual search ability (particularly the divided attention condition) correlated with prospective memory task performance and especially with performance on the second prospective memory task (the reading task). These correlations were more pronounced for the older participants. The explicit retrospective memory tests showed somewhat similar patterns of correlations, especially the cued recall test, which is reasonable given the patterns of correlations found between prospective and explicit retrospective memory task performance. Second, verbal ability (as measured by the NAART) was positively 159 correlated with prospective memory scores for the older group, but uncorrelated for the younger group. However, when the variability due to age was controlled for, positive correlations resulted. Suggesting further similarities to prospective memory performance, the explicit retrospective memory task scores also correlated positively with verbal ability. Finally, the WBSI and the DES were found to correlate significantly and positively with aspects of self-rated prospective memory ability, indicating that, in general, higher levels of self-reported prospective memory forgetting and cue use are related to a tendency toward dissociation and thought suppression. These general patterns of correlations remained when partial correlations controlling for age were computed (see Tables 17 and 18). Results Discussion The results of the research study conducted for this dissertation provide evidence to suggest that prospective memory retrieval is likely not a distinct form of memory and is likely a specific aspect of explicit, episodic memory, as are free recall, cued recall, and recognition. First of all, no functional dissociations between prospective and retrospective tests were found with respect to age. Younger participants scored significantly higher than did older participants on all the memory tests administered, including the implicit test. Although a significant age effect was not expected for implicit word completion, slight age differences on tests of implicit memory are not uncommon. For example, Light and Singh (1987) pointed out in each of their experiments that, although not statistically significant, young participants consistently had a higher rate of word completion than did the older participants, and other researchers have found significant age differences as well (e.g., Chiarello & Hoyer, 1988; Habib, Jelicic, & Craik, 1996). Dividing the older adults into two groups of 160 participants representing "old" (63-75 yrs.) and "oldest" (76-92 yrs.) age groups did not change the pattern of results substantially, but the age effect for implicit word completion was no longer statistically significant for the young and old groups. Age difference effect sizes were higher and more comparable for the prospective and explicit tests than for implicit word completion. However, the pattern of effect sizes was not quite consistent with Craik's (1986) postulated hierarchy of memory tasks, which predicts prospective memory tests should show the largest deficits due to age given a higher dependence upon self-initiated processing than other tests of memory. Secondly, although the levels of processing manipulation produced dissociations between the implicit test and the two explicit tests, as expected, its effect on the prospective memory test was much less clear. Contrary to previous research that found levels effects on prospective memory tests (e.g., Einstein et al., 1995; Meier & Graf, 2000; West & Craik, 1999), the results here indicated no effect of the levels manipulation on prospective memory test performance (see Darby & Maylor, 1998). These findings suggest a dissociation between memory performance on the prospective memory test and memory performance on the explicit cued recall and word recognition tests. In addition, they suggest that prospective memory performance may resemble implicit memory performance more closely than explicit, a finding that would support the idea that unprompted cue recognition occurs spontaneously and automatically (see Einstein & McDaniel, 1996). However, as stated earlier, such conclusions must be made with caution as the levels manipulation for the prospective memory tasks was weak (only one target cue word per study orientation and only applied to the first prospective memory test). Furthermore, when prospective memory target study words were collapsed across study condition (semantic and nonsemantic words combined) and compared with nonstudied target words, no benefit of previous 161 study/ exposure was found for probability of responding to prospective cue words. At first glance, this seems surprising. However, given that the actual prospective memory target was the three-letter combination "res" at the beginning of words and word puzzles, it seems likely that participants were responding to those three letters rather than to entire words. In fact, during testing it was observed that when the cues were embedded within the word completion and cued recall tests, many participants responded to the cue "res" (i.e., they spelled out the three letters r-e-s) before they even completed the word stem or fragment as well as when they could not think of a possible completion. Thirdly, the relationship between prospective and retrospective memory tasks was assessed further by calculating partial correlations between the memory tests, controlling for the influence of age. The results clearly indicated that scores on the prospective memory tasks were positively correlated with scores on the explicit memory tasks, but showed no association with the implicit test. Specifically, both prospective memory tests correlated significantly with cued recall, and the first prospective memory test correlated with word recognition as well (rs = .32 to .35). Normal dissociations were observed between the prospective and word completion tests, suggesting that performance on the tests is independent. A similar pattern of intercorrelations was found for both the young and older adults when zero-order correlations were computed separately for each group. Again, prospective memory performance did not correlate with implicit word completion. Interestingly, prospective memory test 1 scores were not correlated with cued recall for the young group, but were significantly correlated for the old group. Cued recall scores were correlated with prospective memory task 2 scores for both age groups, and word recognition was correlated with prospective memory task 1 scores. What might account for this 162 difference? One would assume that since prospective memory task 2 (the reading task) involved reading a story and having to notice all words beginning with "res" within the story as prospective cues, performance would be more likely to correlate with performance on the word recognition task. However, it may be the case that the reading task places greater demands on self-initiated processing than prospective memory task 1, and as a result be more closely related to cued recall, which depends upon self-initiated processing to a greater extent than recognition memory tests (Craik, 1986). Drawing from the frontal lobe hypothesis of prospective memory, Glisky (1996) predicted that (1) most standard tests of retrospective memory will not correlate with tests of prospective memory, especially when the retrospective component in the prospective task is minimal, and (2) prospective and retrospective tasks that both rely significantly on frontal lobe functioning should correlate. The present findings do not support the first prediction as correlations were found even though the retrospective component of the prospective memory tasks was very simple and careful attention was paid to making sure participants understood exactly what was to be done and when to do it. The second prediction is difficult to address as determining the extent to which tasks rely on frontal lobe functioning and/or self-initiated processing can be quite difficult and is often circular. However, the correlations with cued recall are consistent. Burgess and Shallice (1997) have argued that prospective memory task performance necessarily relies on retrospective remembering and involves a special application of the processes used in episodic retrospective remembering. They proposed that prospective and retrospective memory tasks should not be doubly dissociable; only a single dissociation between prospective and retrospective tasks should ever be observed. In other words, individuals showing significant impairments in retrospective 163 memory abilities should always perform poorly on prospective memory tasks, but individuals with problems performing prospective tasks will not always have retrospective memory difficulties. They also argued that relationships between prospective and retrospective remembering are most likely to be observed when studying individuals with a marked impairment in retrospective memory functioning. Although the older participants in this study showed significantly poorer performance than did the younger participants on the retrospective memory tests, their scores did not demonstrate marked impairment. Nevertheless, for the older group, cued recall performance correlated significantly with performance on prospective memory task 1 (r = .49) whereas these tests were not related for the younger group (r = .20), and the correlation between cued recall and prospective memory task 2 was slightly higher for the older participants than it was for the younger participants. As discussed previously, empirical research in the area of prospective memory is associated with a number of methodological challenges. Problems with testing and measuring prospective memory retrieval have been the root of much of the inconsistency in the research findings. The present study included two event-based tasks of episodic prospective memory in order to assess the extent to which different tasks actually measure the same phenomenon-prospective remembering. One task (the reading task) was adapted from previous research (Kvavilashvili, 1998) and the other was designed specifically for the purposes of the present study. An attempt was made to match these tasks on a number of features. For example, both tasks were repeated intention designs, included the same prospective memory cue ("res" at the beginning of words), included the same number of cues, had retention intervals of approximately the same length (10 minutes), and required participants to make the same response (spell out words associated with the target cue). However, the cues in 164 the first task were embedded within the word completion, cued recall, and word recognition tests, whereas the cues in the second task were embedded within a short story that participants were instructed to read aloud. Although both tasks showed similar effects due to age and levels of processing, the correlation between them was not as large as was expected (r = .38, with the effects of age partialled out). In fact, there was no significant correlation between the two tasks for the young group, whereas the older group showed a significant positive correlation (r = .55). In addition, both the younger and older groups scored higher in the second prospective memory task than they did in the first task. However, this result may be an artifact of testing order. Furthermore, prospective memory task 2 produced a greater number of ceiling effects than did prospective memory task 1. Before possible reasons for these performance differences between the two prospective memory tests are discussed, it is important to note that this research is largely exploratory and the reliability of these measures of prospective memory retrieval have not been demonstrated. Therefore, the results should be interpreted as suggestive and informative for guiding future research and developing prospective memory tests. That said, it is still useful to consider potential explanations for the pattern of findings. One reason why performance on the two prospective memory tests differed could be differences in the flow of the ongoing tasks. As the first prospective task occurred over the course of the three retrospective tasks, slight breaks in prospective memory task performance occurred when switching from one retrospective task to the other and administering task instructions. By contrast, once the reading task began, participants did not pause or switch attention until the task was completed. It also could be the case that when the ongoing task consists of multiple activities (e.g., the three retrospective memory tasks) rather than a single activity (e.g., reading the story), 165 it might result in a higher level of absorption in the ongoing task. In other words, changing activity periodically may help to maintain interest and focus on the ongoing task. Such attention switching requirements and high level of task absorption could place higher demands on available cognitive resources and interfere with prospective memory retrieval to a greater extent than when attention switching was not required and level of absorption in the ongoing task is lower. McDaniel and Einstein (2000b) report results from a study that included both single activity and multiple activity ongoing tasks. The single activity ongoing task condition produced significantly higher prospective memory performance than did the multiple activity ongoing task condition. Therefore, consistent with these findings, the higher levels of performance observed for both younger and older adults on the second prospective memory test (single activity) compared to the first test (multiple activity) may be more a function of the differing demands of the ongoing tasks in each test than a result of the order in which the tests were administered. It also is possible that memory performance could have been influenced by both factors. Thus, even when prospective memory tasks appear to have many characteristics in common, they may not necessarily have the same requirements or may not even tap into many of the same mental processes. Similar to retrospective memory, some prospective memory retrieval tasks may be more demanding of effortful, strategic, self-initiated processes, whereas others may depend primarily on more involuntary, automatic, environmentally-driven processes. Some researchers have postulated that a major difference between prospective and retrospective memory tests is whether participants are in "retrieval mode" (Tulving, 1983) at the time when memory retrieval is supposed to take place (e.g., Einstein & McDaniel, 1996; McDaniel, 1995; McDaniel & Einstein, 2000a). As outlined above, retrospective memory test instructions induce or direct subjects into a retrieval mode, 166 which initiates remembering and sensitizes subjects to the meaning of retrieved memories. By contrast, prospective memory tests require subjects to notice environmental cues as signs of a previously formed intention that must, in turn, trigger memory for the content of the intention-in the absence of being in a retrieval mode. In addition, retrieval modes can vary across memory tests. For example, the retrieval modes for explicit and implicit tests of memory are quite different, with explicit memory retrieval involving a conscious, deliberate attempt to recollect information from a specific prior episode and implicit retrieval lacking any awareness or intentional search of prior episodes. Potential effects of "retrieval mode" were investigated in association with the first prospective memory test, and performance was found to vary substantially, with the word completion context associated with the highest prospective memory response rates, followed by the cued recall context, and the word recognition context with the lowest response rate. It seems somewhat contradictory that the one context in which participants were not actually in a retrieval mode resulted in the highest level of prospective remembering. However, given that implicit memory retrieval is largely environmentally-driven, a greater pool of attentional resources would be available to devote to the prospective memory task than when the retrieval context was either cued recall or recognition, which place greater demands on self-initiated processing. Alternatively, this pattern of results may be due to differences in the extent to which the ongoing task involved focal processing of the prospective memory cue (e.g., Meier & Graf, 2000). If the prospective memory cue is encountered, but it is not part of the information being attended to or processed for performance of the ongoing task, it may be less likely that it will be noticed in that capacity. Given that noticing the cue "res" requires processing the physical features of words rather than processing the meaning of words, it makes sense that the highest level of prospective memory 167 performance was observed in the word completion task, which is a data-driven implicit memory test. Participants focus on the arrangement of letters in the word stems and fragments in order to generate words that fit those constraints. By contrast, when performing the cued recall and recognition tests, participants are much more likely to be focused on the meaning of words rather than their physical attributes in attempting to remember if they were presented in the previous study episode. The fact that word recognition resulted in the lowest prospective memory response rate may have to do with the additional task demands associated with the presence of distractors that were designed to be quite similar to target words. Finally, it is possible that these retrieval context findings simply reflect a test order effect. However, in a repeated intentions design one would expect performance to increase or at least remain fairly stable as a result of practice, especially with a relatively small number of cues (e.g., Ellis et al., 1999; Maylor, 1993a). In the first prospective memory test, performance decreased across trials and no practice or order effects were observed across trials of the second prospective memory test. In addition to the objective measures of prospective memory retrieval, self-perceptions of prospective ability were explored and compared to performance on the objective prospective memory tests. Consistent with previous research (see Maylor, 1996b), older adults rated themselves as less forgetful than younger adults in remembering intentions in everyday life. Younger adults reported more failures to carry out both long-term episodic and short-term habitual prospective memory tasks. Younger adults also scored higher in usage of internal cues and external techniques to remember. Interestingly scores on the PMQ did not correlate with performance on either of the laboratory-based prospective memory tests. The lack of correlation was applicable for both young and old adults and when the variability due to age was 168 partialled out. Again, the failure to find an association between self-rated memory functioning and scores on memory tests is not unusual (Herrmann, 1984; Kausler, 1994). By contrast, self-assessments of prospective memory performance on the laboratory tasks and actual performance correlated highly for both the younger and older adults. These findings are consistent with those of Hertzog et al. (2000) who found that self-ratings on meta-memory scales did not predict medication adherence, but asking participants whether they remembered to take their medication did. It seems likely that the differences in the self-assessments may have more to do with differences between carrying out prospective memory tasks in the real world where numerous factors can affect whether an intention is remembered and/or carried out (e.g., external reminders, importance of the intention, unexpected events) versus prospective retrieval in a controlled laboratory setting than with factors associated with meta-cognitive aspects of prospective remembering (see Birt, 1999; Maylor, 1996b). Although self-reported prospective memory performance was not related to memory performance on the objective tests, it was related to a number of individual difference factors including scores on the WBSI and the DES. In general, as hypothesized, tendency to suppress thoughts and tendency toward dissociation were associated with higher levels of forgetting on the PMQ as well as a greater tendency to rely on both internal and external cues in order to remember. Thus, individuals who are prone to rumination and obsessiveness, more in touch with external reality, and not as easily distracted are more likely to report success in remembering to carry out intentions in everyday life. With respect to performance on the two prospective memory tasks, a number of individual difference factors were found to be correlated, the most obvious of which were visual search and attention ability (especially visual search while dividing attention) and verbal ability. Visual search and verbal ability 169 scores also were correlated with retrospective cued recall and recognition (but not implicit word completion), further demonstrating similarities between episodic prospective and retrospective memory retrieval. Three theoretical views proposed to describe prospective memory retrieval (i.e., unprompted cue recognition) were reviewed: the recognition-recall view, the automatic associative activation view, and a functional view of memory. The recognition-recall, or familiarity plus search view, assumes that processes mediating recognition memory can fully account for prospective memory (e.g., Einstein & McDaniel, 1996; McDaniel, 1995). This model posits that prospective memory retrieval involves two distinct sequential processes: a relatively automatic cue recognition memory process followed by a purposeful and controlled search of memory. According to this model, although both memory processes are involved, performance on prospective memory tests should more closely resemble performance on recognition memory tests than tests of recall. By contrast, the automatic associative activation model is based on the assumption that during the encoding phase of prospective memory tasks, an association between the target cue and the to-be-performed action is formed (see Einstein & McDaniel, 1996; Goschke & Kuhl, 1993, 1996; McDaniel & Einstein, 1993; McDaniel et al., 1998). Whether the target cue will be recognized as a sign of a previously formed intention depends on the activation level of the cue-action association at the time the target retrieval cue occurs. According to this model, prospective remembering occurs automatically. No directed, purposeful search of memory is required. When the activation level reaches its threshold, an encounter with the target cue automatically triggers the activation of its associated action. This model posits that prospective memory retrieval should more closely resemble the type of remembering that occurs in implicit memory tasks, which (are assumed to be) are 170 carried out by automatic, involuntary processes. The functional view of memory maintains that memory test performance is determined by the extent to which it depends upon the availability of processing resources. Craik (1983, 1986) proposed that memory tests can be arranged along a continuum, according to the extent to which performance depends on subject-initiated processes that are effortful and controlled at one end environmentally-driven processes that are guided primarily by retrieval cues at the other end. Due to the fact that prospective memory retrieval requires unprompted cue recognition, Craik argued that it relies primarily upon self-initiated processes. In fact, according to his proposed hierarchy of memory tasks, prospective memory is more dependent upon self-initiated processes than free recall, cued recall, and recognition. If the functional account of prospective remembering is accurate, prospective memory performance should show more similarities with recall tests than recognition tests, and even fewer similarities with implicit tests of memory. The results of the present research indicate clearly that prospective memory performance is not related to implicit memory test performance, but more closely resembles performance on explicit memory tests. However, the findings with respect to explicit memory tests are not so clear-cut. Prospective memory retrieval was found to be related to both word recognition and cued recall. Specifically, prospective memory task 1 performance was related to performance on the recognition as well as cued recall test. By contrast, prospective memory test 2 scores were related largely to scores on cued recall, but not recognition. In addition when the pattern of young-old age difference effect sizes for the various tests of memory administered were examined, the prospective memory test 1 (d = .64) and cued recall (d = .69) effect sizes were closer in magnitude than were prospective memory test 2 (d = .86) and cued recall. However, the differences were not very large and were much more similar 171 to one another than to the effect size for word recognition (d = 1.39). This pattern of age-related differences in memory supports the findings from the meta-analysis of the research literature, which revealed that age-related prospective memory task performance declines appear to be comparable to the declines that have been observed with explicit, episodic memory tasks. Thus, little support was found for the automatic associative view of prospective memory, and although some evidence was uncovered for the recognition-recall view of prospective memory, the overall pattern of findings more strongly support the functional view of memory. That said, it is recognized that similar to retrospective memory tasks, some prospective memory retrieval tasks may rely on effortful, self-initiated processing more heavily than others, whereas some may depend largely on spontaneous, automatic processes (e.g., Gollwitzer, 1999; McDaniel & Einstein, 2000b). While the two prospective memory tasks included in this study correlated only modestly (r = .38) and differed in a number of important ways, it was apparent that the key similarity between the tasks -unprompted cue recognition of a previously formed intention-shares many of the same mental processes underlying explicit, episodic memory retrieval. 172 GENERAL DISCUSSION To summarize, the primary objectives of this research were: (1) to test the claim that prospective memory retrieval is functionally distinct, and (2) to investigate age-related differences in prospective memory test performance and their relation to age differences in different types of retrospective memory tests. Implicit in the term "prospective memory" is the assumption that it is a unique type of memory, distinct and dissociable from the various forms of retrospective memory (e.g., explicit, implicit; episodic, semantic, procedural). This dissertation provides the first attempt to test this assumption directly. At the outset, a comprehensive examination and review of the literature on prospective memory was presented. A number of the key issues were highlighted, including problems with defining and operationalizing prospective memory, methodological concerns about testing and measuring prospective memory retrieval, theoretical models and frameworks proposed for understanding prospective memory performance, and inconsistencies in the research findings to date. The variety of ways in which prospective memory tasks can differ was discussed and following a detailed task analysis highlighting the difference between tasks and task components, the importance of focusing on the prospective memory component of prospective memory tasks was emphasized. Recognizing that many task features and characteristics can influence prospective memory performance, the scope of inquiry was restricted primarily to event-based explicit episodic prospective memory retrieval. After considering a number of ways in which prospective and retrospective memory retrieval might differ, it was proposed that the key, defining feature of prospective memory retrieval is unprompted cue recognition. Unprompted cue recognition was defined here as the recognition of a naturally occurring cue embedded within one's ongoing 173 activities and environment (internal or external) as a sign of a previously formed intention in the absence of any external prompts or focus on remembering prior experiences. Unlike retrospective memory retrieval in which the rememberer is in a retrieval mode for remembering previous experiences and sensitized for cue recognition, prospective memory cue recognition occurs in the absence of a retrieval mode and attention is usually directed at some other ongoing activity. If the retrieval cue is successful in triggering memory for an intention (i.e., the anticipatory state of knowing that something must be done), it must then lead to the initiation of retrieval processes that will hopefully result in remembering exactly what was intended. As discussed previously, when the association between the intention and its content is weak or is temporarily blocked resulting in a failure to remember what was originally intended, tip-of-of-the-tongue experiences and feeling of knowing states similar to those that can be experienced with retrospective memory retrieval failures can result. However, when the association is strong and the retrieval cues/ context is rich, the content of the intention is remembered easily (e.g., Einstein & McDaniel, 1996; Gollwitzer, 1999). In order to summarize the findings in the research literature concerning the relationship between prospective and retrospective memory performance, a quantitative review of all available empirical studies including prospective-retrospective task comparisons was conducted. This review revealed that, at present, there is little convincing evidence from patterns of both functional and stochastic dissociations to support the argument that prospective memory is a functionally distinct activity or system of memory. Experimental, neuropsychological, and developmental evidence suggest as many similarities between prospective and retrospective tasks as differences. Furthermore, the meta-analysis conducted on the effects of aging found 174 the magnitude of age differences in prospective memory performance to be similar to that found in episodic retrospective memory task performance (see Birt, 1999). It also was found that the magnitude of age differences varied substantially according to a number of prospective memory task features and characteristics. In an attempt to more clearly delineate the relationship between prospective and retrospective memory tasks, and provide insight into whether the mental processes required for remembering to do things in the future are similar to those required for remembering experiences from the past, a controlled laboratory-based research study was conducted comparing performance on retrospective word completion, cued recall, word recognition, and two event-based prospective memory tasks. The memory tests were matched on as many test properties as possible in order to isolate memory retrieval processes, reduce confounding variability, and facilitate direct comparisons. In addition, the influence of the variables age and levels of processing on prospective and retrospective memory tasks were explored to test for possible functional dissociations. Together, the results from the two quantitative literature reviews and the empirical research study indicated that the memory processes required for episodic prospective memory retrieval are much more similar to those required for explicit episodic retrospective memory retrieval than for implicit retrospective memory. Thus episodic prospective memory retrieval does not appear to be functionally distinct from episodic retrospective memory retrieval. These results offer support to theoretical views that assume prospective memory depends primarily on effortful, self-initiated processes (e.g., Craik, 1986), but are not consistent with views that maintain prospective retrieval is largely automatic (e.g., Einstein & McDaniel, 1996). If episodic prospective memory performance is not functionally distinct from episodic retrospective memory performance, and episodic retrospective memory 175 performance is known to consist of a number of different activities or tasks of memory (e.g., free recall, cued recall, recognition), where might prospective memory retrieval fit within this current taxonomy of memory tasks? Based on Craik's (1986) hierarchy of memory, which places memory tasks on a continuum according to the degree to which they require self-initiated versus environmentally-driven processing, prospective memory tasks are assumed to be at the top of the hierarchy, demanding self-initiated processes to the greatest extent. However, in this study, prospective memory performance was related to both cued recall and recognition memory performance. Although the patterns of correlations differed somewhat between the two prospective memory tests administered here, there was a trend for a slightly stronger relationship between prospective memory retrieval and cued recall than between prospective memory and recognition (e.g., both prospective tests were related to cued recall, but only the first test was related to recognition). This finding is consistent with the findings from the quantitative review of the literature. When stochastic dissociations were examined, cued recall tests were less likely than recognition tests to show performance dissociations with tests of prospective memory. This combination of results may be taken as evidence to suggest that the mental processes underlying prospective memory retrieval more closely resemble those that underlie cued recall than recognition. However, these results do not mean that the processes are the same. Given that the correlations between prospective memory and cued recall performance were not as strong as might be expected and because the task analysis revealed distinct features of prospective memory retrieval, it also is concluded that prospective memory retrieval can be considered a unique aspect or activity of episodic memory, in the same way that recall and recognition are unique memory activities. 176 Several limitations of this research should be noted, but before they are addressed it is important to emphasize the exploratory nature of this research. As stated earlier, research on prospective memory is relatively new and methodologies and experimental paradigms continue to be developed and refined. Many topics and issues concerning prospective memory have yet to be explored. In the present study, in addition to the primary questions examined, many exploratory statistical analyses were conducted. Therefore, the possibility of an inflated type 1 error rate must be acknowledged. However, given that this research was not designed to be definitive but to serve as an in-depth exploration into prospective memory retrieval, the potential for type 1 error inflation should not be regarded as particularly problematic. It is hoped that the findings of this research will aid in initiating and guiding fruitful avenues for future research. The first limitation of this research concerns the specific tests selected to measure both prospective and retrospective memory retrieval. With respect to the implicit memory test and the lack of reliability often found with implicit measures of memory, a replication and extension of the implicit memory findings will be important. Furthermore, the findings with regard to age and retrieval context effects would have benefited from the inclusion of a second, more conceptually-driven implicit memory test. Implicit memory tests can vary according to the extent to which they require data-driven versus conceptually-driven processes (see Roediger & McDermott, 1993). Performance on data-driven tests is primarily initiated or driven by testing materials and sensory processes (bottom-up processing), whereas performance on conceptually-driven tests is largely initiated and driven by thought and meaning (top-down processing). Word stem and fragment completion are considered to be data-driven implicit tests, whereas implicit tests such as category production and free association 177 are considered to be more conceptually-driven. Examining the patterns of functional dissociations due to age and exploring the likelihood of recognizing the prospective memory target cue in the context of a conceptually-driven implicit test would help to clarify the findings concerning the nature of the relationship between prospective and implicit memory retrieval. In addition, the use of a within-subjects design may have influenced performance on the memory tests. Therefore, replicating the pattern of relationships between performance on the prospective and retrospective memory tests using a between-subjects design would offer convincing support for the findings. Including a test of retrospective free recall also would help to delineate more specifically where episodic prospective memory retrieval fits within current classifications of memory tests. Because research into prospective memory is relatively new and experimental paradigms, methodologies, and specific tests for measuring prospective memory retrieval continue to be developed, investigating patterns of dissociations between prospective and retrospective memory performance using a larger variety of prospective memory tests (more than two) would have been very informative. For example, it would be interesting to see how prospective memory performance on the frequently-used computer-based working memory prospective memory test (e.g., Einstein & McDaniel, 1990) compares to the memory performance on the two prospective memory tests included in this study. Also, patterns of dissociations should be investigated with time-based prospective memory tasks, which, although similar to event-based tasks, may show an even stronger relation to episodic retrospective memory tasks such as cued and free recall given the postulated high demands time-based tasks place on self-initiated processing. Although the correlations between scores on the two tests of prospective memory were modest, they indicate a need for 178 future research focused on developing valid and reliable measures of prospective memory retrieval. Especially important, not to mention challenging, is devising ways to isolate and measure prospective memory retrieval more closely. Although a combination of methods were used in this study in an attempt to measure prospective memory retrieval as directly as possible, overall task performance was still measured to an extent. Once the prospective memory component of prospective memory tasks can be measured directly, then a much better understanding of how it is similar to and unique from retrospective memory retrieval can be obtained. Another limitation of the research conducted for this dissertation concerns the effects of age on prospective memory retrieval. The older adult age group included participants of a wide range of ages (63 - 92 years). It is likely that, in general, the cognitive abilities of 60- and 70-year-olds are different from those of 80- and 90-year-olds and that many individual differences exist between such age groups. Researchers interested in the effects of age on prospective memory retrieval have recognized the need to divide older research participants into specific age bands (e.g., Huppert, Johnson, & Nickson, 2000; Mantyla & Nilsson, 1997), which allows for a much finer grained approach to studying age-related deficits in memory retrieval, reduces confounding variability in performance differences, and results in a more accurate account of the progression of declines in prospective memory functioning with age. Although the older age group was divided into 'young-old' and 'old-old' groups for analyses of age effects in this study, it is likely these analyses lacked statistical power to detect group differences. Of course, research employing longitudinal designs would offer the most accurate account of patterns of decline accompanied by increasing age. In addition to normal aging, further research on prospective memory retrieval needs to be conducted with individuals suffering from age-associated memory impairment and 179 mild forms of dementia. Although little research on prospective memory has been conducted with such subject populations, one research study (Huppert & Beardsall, 1993) suggested that prospective memory impairments may be an early warning sign or indicator of dementia. As the proportion of elderly persons within the population continues to rise and the prevalence of dementia increases, the development of tests for detecting dementia in the early stages of the disease and differentiating between changes in memory that are a result of normal aging versus those that are indicative of a dementia such as Alzheimer's disease have become increasingly important (Huppert et al., 2000). Thus, exploring the utility of prospective memory tests in the early diagnosis of Alzheimer's disease may represent an important avenue for future research. Although the focus of this research was on prospective memory retrieval, retrieval processes, of course, do not occur in isolation. Memory performance is multi-faceted and is the result of a complex set of interactions between a particular rememberer (with specific cognitive abilities, beliefs, and a prior history of experiences) and a particular environmental context (e.g., specific memory tasks) with a unique set of characteristics and demands. The probability with which prospective memory is likely to be successful can be viewed as the dynamic interaction between features of the prospective memory task, the target cue, the ongoing task, and the individual. As discussed previously, a number of researchers have identified factors, other than failures in prospective memory retrieval, that can influence whether prospective memory tasks are actually carried out, especially outside the laboratory in real-world contexts (e.g., Dobbs & Reeves, 1996; Einstein & McDaniel, 2000b; Ellis, 1996; Meacham, 1982, 1988; Park & Kidder, 1996; Winograd, 1988). Such factors include, but are not limited to, aspects related to goal-setting, planning, attention, monitoring, 180 importance of the intention, self- versus other-generated intentions, retrieval cue properties, retention interval, features of the ongoing activity, problem-solving, motivation, compliance, reward, interpersonal relations, cognitive abilities, personality, affect, meta-cognitive beliefs, reality monitoring, output evaluation, etc. Although many of these factors have been demonstrated to be important for whether information and events are remembered from the past, some may be more relevant in determining whether intentions are remembered and carried out in a timely fashion. Several of these factors are discussed to highlight their importance in determining prospective remembering, especially with respect to real-world settings and to suggest areas for future research. An obvious task characteristic that influences prospective memory retrieval is the importance of the to-be-remembered intention. The more important a prospective memory task is perceived to be, the more likely it is to be remembered and carried out (e.g., Andrzejewski et al., 1991; Cicogna & Nigra, 1998; Kvavilashvili, 1987; Meacham & Singer, 1977). Although only a few studies have investigated this relationship empirically, the findings suggest a positive relationship with prospective memory retrieval and a negative relationship with performance of the ongoing activity (see McDaniel & Einstein, 2000b). McDaniel and Einstein (2000b) suggested that when prospective tasks are important, individuals are more likely to devote effort and strategic processes in an attempt to ensure that the intention is not forgotten (e.g., engaging in more frequent mental rehearsals, setting up external reminders and retrieval cues). By contrast, when prospective tasks are less important, individuals are likely to not bother exerting the mental effort and simply leave prospective memory retrieval to more automatic and spontaneous processes. Of course, employing strategies to help prevent prospective memory failures does not always ensure that 181 such failures will not occur. For example, one may get distracted by ongoing activities and not recognize retrieval cues as signs of the intention when they are encountered. In addition, it seems that deliberate, strategic processing is often not required in cases when the intention is perceived to be very important, is relatively novel, and associated with significant consequences. Under these circumstances, the rememberer is often confident that the importance of the intention will ensure that it will be remembered successfully without having to set up external cues or reminders. Perhaps such confidence is due to the meta-knowledge that because of their significance, there is a natural tendency to mentally rehearse important intentions and to associate them with a wide variety of retrieval cues. In other words, similar to what has been found with retrospective memory (e.g., Craik & Lockhart, 1972), information perceived to be important and personally-relevant tends to be processed more deeply and elaborately than less important information, which leads to a greater likelihood that subsequent memory retrieval will be successful. A factor closely related to the perceived importance of the intention that is likely to influence prospective remembering concerns the source of the intention. Intentions can differ according to whether they are formed as a result of a personal need or desire to do something (self-generated) or whether they stem from a request from someone else (other-generated) (Cohen, 1989; Ellis & Nimmo-Smith, 1993; Kvavilashvili & Ellis, 1996). Memory research has demonstrated that personally relevant information tends to be remembered better than information that is more difficult to relate to the self. In addition, information that is self-generated tends to be remembered better than information that is provided by others-a finding referred to as the generation effect (e.g., Slamecka & Graf, 1978). Thus, one would expect to find a similar memorial advantage for self-generated intentions over other-generated 182 intentions. Given that experimental prospective memory tasks have thus far employed only experimenter-provided intentions, does this limit the generalizability of the research findings to real world settings? Probably not. In many cases other-generated intentions can be considered equally (or even more) important as self-generated intentions. Even though an intention may have originated from another person, it may become highly self-relevant and may be processed just as elaborately as if it had been originated from oneself in the first place. Thus, although the generation effect has yet to be investigated within the context of prospective memory retrieval, one would expect potential interactions with related factors such as importance, compliance, and depth of processing. Two additional phenomena that have been discussed as relevant to prospective memory performance, especially with respect to the execution of intentions, are reality monitoring and output monitoring. Reality monitoring refers to processes required for distinguishing between memories for events that were simply thought about and imagined versus events that were actually perceived and experienced in reality (Johnson & Raye, 1981). Reality monitoring is applicable to performance in prospective memory tasks in that one often has to make decisions regarding whether a previously formed intention was simply contemplated or whether it was actually carried out. Thus, the memory for an intention can become confused with the memory for performing that intention. Successful reality monitoring is important for preventing failures to carry out or execute intentions (Maylor, 1996b). Output monitoring refers to processes concerned with distinguishing between those intentions that have been performed and those that still need to be carried out (Koriat & Ben-Zur, 1988). Output monitoring is important for remembering that intentions have been realized and to prevent performance repetitions. There is some evidence to suggest that the elderly 183 are more likely than the young to forget that a planned action has already been performed and therefore repeat it (Koriat, Ben-Zur, & Sheffer, 1988). Again, underscoring the complexity of the dynamics of prospective memory tasks, it seems likely that success in reality and output monitoring of intentions would be related to the importance of the intention, whether it was self- or other-generated, compliance, and depth of processing. Although not exhaustive, discussion of the relevance of factors such as perceived importance of the intention, the generation effect, depth of processing, reality monitoring and output monitoring to prospective memory performance underscores the complexity of prospective memory tasks. It also reiterates the point that, although the term prospective memory is often used as if it refers to a single manner in which memory can be expressed, it is clear that prospective memory retrieval can occur within many different types of memory tasks (e.g., explicit, implicit; episodic, habitual; short-term, long-term), all of which can differ on numerous task properties (e.g., type of study, retention interval, retrieval cues, context, environment, etc.). The research presented in this dissertation focused on explicit episodic event-based prospective memory retrieval and revealed some important findings and insights with respect to its relationship to explicit episodic and implicit retrospective memory retrieval and the magnitude of decline in performance that can be expected as a function of normal aging. These results have potential implications for informing and improving upon current theoretical models of prospective memory retrieval as well as theories of aging and memory. They also will help to improve methodological approaches for testing and evaluating prospective memory retrieval and, given that the specific processes involved in a prospective memory situation are likely to vary across tasks and individuals, the results increase the awareness of the importance of studying 184 individual difference factors and how they interact with task demands in explaining the variability in prospective memory performance. It is hoped that this research will spark interest in investigating prospective memory retrieval in many different contexts in order to increase our knowledge of the key underlying processes and influential variables as well as to further our understanding of how prospective memory retrieval fits into our current understanding of memory functioning. 185 Footnotes Vrospective remembering is defined here as unprompted cue recognition; that is, remembering a previously formed intention at some point in the future, in the absence of any external prompting or reminders. Therefore, a number of experimental paradigms and research studies that employ explicit retrieval instructions will be excluded from the present discussion (e.g., Brooks & Gardiner, 1994; Goschke & Kuhl, 1993; Hitch & Ferguson, 1991; Koriat, Ben-Zur, & Nussbaum, 1990; Mastroianni et al., 1996; Sinnott, 1986). Nevertheless, there is some evidence that self-assessment of memory abilities can be quite accurate, especially when the questions are directly related to everyday actions (e.g., Bennett-Levy & Powell, 1980; Martin, 1986). 3 These criteria are not meant to be exhaustive. Many other sets of criteria have been suggested and utilized in arguing for the existence of distinct memory systems (e.g., format for representing information, length of storage time, phylogenetic and ontogenetic development, kinds of consciousness associated with each). The four described here are used as examples because they are representative of some of the more common criteria cited as necessary for specifying separate memory systems. 186 References Anderson, J . R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press. Andrzejewski, S. J . , Moore, C. M., Corvette, M., & Herrmann, D. (1991). Prospective memory skill. Bulletin of the Psvchonomic Society, 29, 304-306. Atkinson, R. C , & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. In K. W. Spence, & J . R. Spence (Eds.), The psychology of learning and motivations: Advances in research and theory (Vol. 2, pp. 89-195). New York, NY: Academy Press. Baddeley, A. D. (1986). Working memory. Oxford: Clarendon Press. Baddeley, A. D., & Wilkins, A. J . (1984). Taking memory out of the laboratory. In J . E. Harris, & P. E. Morris (Eds.), Everyday memory, actions and absent-mindedness. New York, NY: Academic Press. Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology. Cambridge: Cambridge University Press. Beck, A. T., & Steer, R. (1987). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. Bennett-Levy, J . , & Powell, G. E. (1980). The subjective memory questionnaire: An investigation into the self-reporting of real-life memory skills. British Journal of Social and Clinical Psychology, 19, 177-188. Bernstein, E. M., & Putnam, F. W. (1986). Development, reliability, and validity of a dissociation scale. Journal of Nervous and Mental Disease, 174, 727-735. Birt, A. R. (1999). Prospective memory and aging: An exploratory meta-analysis. Manuscript submitted for publication. 187 Bisiacchi, P. S. (1996). The neuropsychological approach in the study of prospective memory. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 297-317). Mahwah, NJ : Erlbaum. Blackburn, A. B. (1995). Involuntary recollection: How does a cue trigger the spontaneous recollection of a memory? Dissertation Abstracts International: Section B: The Sciences and Engineering, 55, 5585. Blair, J . R., & Spreen, O. (1989). Predicting premorbid IQ: A revision of the National Adult Reading Test. The Clinical Neuropsychologist, 3, 129-136. Blaxton, T. A. (1989). Investigating dissociations among memory measures: Support for a transfer appropriate processing framework. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 657-668. Bower, G. H. (1981). Mood and memory. American Psychologist, 36, 129-148. Brandimonte, M. A., Einstein, G. O., & McDaniel, M A. (1996). Prospective memory: Theory and applications. Mahwah, NJ: Lawrence Erlbaum. Brandimonte, M. A., & Passolunghi, M. C. (1994). The effect of cue-familiarity, cue-distinctiveness, and retention interval on prospective remembering. Quarterly Journal of Experimental Psychology, 47A, 565-587. Bransford, J . D., Franks, J . J . , Morris, C. D., & Stein, B. S. (1979). Some general constraints on learning and memory research. In L. S. Cermak, & F. I. M. Craik (Eds.), Levels of processing in human memory (pp. 331-354). Hillsdale, NJ: Lawrence Erlbaum. Brooks, B. M., & Gardiner, J . M. (1994). Age differences in memory for prospective compared with retrospective subject-performed tasks. Memory & Cognition, 22, 27-33. 188 Brown, A. S. (1991). A review of the tip-of-the-tongue experience. Psychological Bulletin. 109, 204-223. Brown, R., & McNeill, D. (1966). The "tip of the tongue" phenomenon. Journal of Verbal Learning and Verbal Behaviour, 5, 325-337. Burgess, P. W. (July, 2000). The cognitive neuroscience of prospective memory. Paper presented at the 1 s t International Conference on Prospective Memory, Hatfield, UK. Burgess, P. W., & Shallice, T. (1996). Response suppression, initiation and strategy use following frontal lobe lesions. Neuropsychologia, 34, 263-273. Burgess, P. W., & Shallice, T. (1997). The relationship between prospective and retrospective memory: Neuropsychologial evidence. In M. A. Conway (Ed.), Cognitive models of memory: Studies in cognition (pp. 247-272). Cambridge, MA: The MIT Press. Burke, D. M., & Light, L. L. (1981). Memory and aging: The role of retrieval processes. Psychological Bulletin, 90, 513-546. Canadian Study of Health and Aging Working Group (1994). Canadian study of health and aging: Study methods and prevalence of dementia. Canadian Medical Association Journal. 150, 899-913. Carlson, E. B., & Putnam, F. W. (1993). An update on the Dissociative Experiences Scale (DES). Dissociation: Progress on the Dissociative Disorders, 6, 16-27. Cavanaugh, J . C. (1989). The importance of awareness in memory aging. In L. W. Poon, D. C. Rubin, & B. A. Wilson (Eds.), Everyday cognition in adulthood and late Nfe (pp. 416-436). Cambridge: Cambridge University Press. 189 Ceci , S. J . , Baker, J . G. , & Bronfenbrenner, U. (1988). Prospective remembering, temporal calibration, and context. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (pp. 360-365). Chichester, NY: J . Wiley & Sons. Cerella, J . (1990). Aging and information-processing rate. In J . E. Birren, & K. W. Schaie (Eds.), Handbook of the psychology of aging, (3rd Ed.) (pp. 201-221). San Diego, CA: Academic Press. Cermak, L. S. (1993). Automatic versus controlled processing and the implicit task performance of amnesic patients. In P. Graf, & M. E. J . Masson (Eds.), Implicit memory: New directions in cognition, development, and neuropsychology (pp. 287-301). Hillsdale, NJ: Lawrence Erlbaum. Cherry, K. E., & LeCompte, D. C. (1999). Age and individual differences influence prospective memory. Psychology and Aging, 14, 60-76. Chiarello, C , & Hoyer, W. J . (1988). Adult age differences in implicit and explicit memory: Time course and encoding effects. Psychology and Aging, 3, 358-366. Cicogna, P. C , & Nigra, G. (1998). Influence of importance of intention on prospective memory performance. Perceptual and Motor Skills, 87, 1387-1392. Clark, D. M., Ball, S., & Pape, D. (1991). An experimental investigation of thought suppression. Behaviour Research and Therapy, 29, 253-257. Cockburn, J . (1995). Task interruption in prospective memory: A frontal lobe function? Cortex, 31, 87-97. Cockburn, J . (1996). Failure of prospective memory after acquired brain damage: Preliminary investigation and suggestions for future directions. Journal of Clinical and Experimental Neuropsychology, 18, 304-309. 190 Cockburn, J . , & Smith, P. T. (1988). Effects of age and intelligence on everyday memory tasks. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 2, pp. 132-136). Chichester, England: Wiley. Cockburn, J . & Smith, P. T. (1991). The relative influence of intelligence and age on everyday memory. Journals of Gerontology 46, 31-36. Cockburn, J . , & Smith, P. T. (1994). Anxiety and errors of prospective memory among elderly people. British Journal of Psychology, 85, 273-282. Cohen, G. (1989). Memory in the real world. Hillsdale, NJ : Lawrence Erlbaum. Cohen, G. (1996). Memory in the real world (2nd ed.). East Sussex, UK: Psychology Press. Cohen, G. , & Faulkner, D. (1984). Memory in old age: 'Good in parts'. New Scientist, 11, 49-51. Cohen, N. J . , & Squire, L. R. (1980). Preserved learning and retention of pattern analysing skill in amnesia: Dissociation of "knowing how" and "knowing that." Science, 210, 207-209. Costermans, J . , & Desmette, D. (1999). A method for describing time-monitoring strategies in a prospective memory setting. Current Psychology of Cognition, 18, 289-306. Craik, F. I. M. (1983). On the transfer of information from temporary to permanent memory. Philosophical Transactions of the Royal Society of London, Series B, 302, 341-359. Craik, F. I. M. (1986). Afunctional account of age differences in memory. In F. Klix, & H. Hagendorf (Eds.), Human memory and cognitive capabilities, mechanisms, and performances (pp. 409-422). North Holland: Elsevier. 191 Craik, F. I. M. (1992). Memory changes in aging. In I. Kostovic, S. Knezevic, H. M. Wisniewski, & G. J . Spilich (Eds.), Neurodevelopment, aging and cognition (pp. 304-320). Boston, MA: Birkhauser. Craik, F. I. M. (1994). Memory changes in normal aging. Current Directions in Psychological Science, 3, 155-158. Craik, F. I. M., & Byrd, M. (1982). Aging and cognitive deficits: The role of attentional resources. In F. I. M. Craik, & S. Trehub (Eds.), Aging and cognitive processes (pp. 191-211). New York, NY: Plenum. Craik, F. I. M., & Jennings, J . M. (1992). Human memory. In F. I. M. Craik, & T. A. Salthouse (Eds.), Handbook of aging and cognition (pp. 51-100). Hillsdale, NJ: Erlbaum. Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671-684. Craik, F. I. M., & Kerr, S. A. (1996). Prospective memory, aging and lapses of intention. In M. Brandimonte, G. O. Einstein, & M.A. McDaniel (Eds.), Prospective Memory: Theory and application (pp. 227-237). Mahwah, NJ : Erlbaum. Craik, F. I. M., & McDowd, J . M. (1987). Age differences in recall and recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 1 ^ 474-479. Crawford, J . R., Stewart, L. E., Cochrane, R. H. B., Parker, D. M., & Besson, J . A. O. (1989). Construct validity of the National Adult Reading Test: A factor analytic study. Personality and Individual Differences, 10, 585-587. Crovitz, H. R., & Daniel, W. F. (1984). Measurements of everyday memory: Toward the prevention of forgetting. Bulletin of the Psychonomic Society, 22, 413-414. 192 Crowder, R. G. (1996). The trouble with prospective memory: A provocation. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 143-147). Mahwah, NJ: Erlbaum. Dalla Barba, G. (1993). Prospective memory: A 'new' memory system? In F. Boiler, & J . Grafman (Eds.), Handbook of Neuropsychology. (Vol. 8, pp. 239-251). New York, NY: Elsevier. Darby, R. J . , & Maylor, E. A. (1998, April). Effects of the relationship between background and prospective memory task requirements on age differences in prospective memory. Poster presented at the Seventh Cognitive Aging Conference, Atlanta, Georgia. Davies, G. M., & Thomson, D. M. (Eds.). (1988). Memory in context: Context in memory. Chichester, England: John Wiley & Sons. Davies, M. I., & Clark, D. M. (1998). Thought suppression produces rebound effect with analogue post-traumatic intrusions. Behaviour Research and Therapy. 36. 571-582. Delbecq-Derouesne, J . , & Beauvois, M. F. (1989). Memory processes and aging: A defect of automatic rather than controlled processes? Archives of Gerontology and Geriatrics. Supplement 1. 121-150. de Wall , C , Wilson, B. A., & Baddeley, A. D. (1994). The Extended Rivermead Behavioural Memory Test: A measure of everyday memory performance in normal adults. Memory. 2. 149-166. Dixon, R. A. (1989). Questionnaire research on metamemory and aging: Issues of structure and function. In L. W. Poon, D. C. Rubin, & B. A. Wilson (Eds.), Everyday cognition in adulthood and late life (pp. 394-415). New York, NY: Cambridge University Press. 193 Dixon, R. A., & Hultsch, D. R. (1983). Structure and development of metamemory in adulthood. Journal of Gerontology, 38, 682-688. .. Dobbs, A. R., & Reeves, M. B. (1996). Prospective memory: More than memory. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 199-225). Mahwah, NJ : Erlbaum. Dobbs, A. R., & Rule, B. G. (1987). Prospective memory and self-reports of memory abilities in older adults. Canadian Journal of Psychology, 41, 209-222. d'Ydewalle, G. (1995). Age-related interference of intervening activities in a prospective memory task. Psychologica Belgica, 35, 189-203. d'Ydewalle, G. , Luwel, K., & Brunfaut, E. (1999). The importance of on-going concurrent activities as a function of age in time- and event-based prospective memory. European Journal of Cognitive Psychology, 11, 219-237. Ebbinghaus, H. (1885/1964). Uber das Gedachtnis. Lepzig: Dunker. Translation by H. Ruyer, & C. E. Bussenius (1913), Memory. New York, NY: Teachers College, Columbia University. Eich, E. (1980). The cue-dependent nature of state-dependent retrieval. Memory & Cognition, 8, 157-173. Eich, E. (1989). Theoretical issues in state dependent memory. In H. Roediger, & F. Craik (Eds.), Varieties of memory and consciousness: Essays in honour of Endel Tulving (pp. 331-354). Hillsdale, NJ: Lawerence Erlbaum Associates. Eich, E., & Metcalfe, J . (1989). Mood dependent memory for internal versus external events. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 443-455. 194 Einstein, G. 0 . , Holland, L. J . , McDaniel, M. A., & Guynn, M. J . (1992). Age related deficits in prospective memory: The influence of task complexity. Psychology and Aging. 7. 471-478. Einstein, G. O., & McDaniel, M. A. (1990). Normal aging and prospective memory. Journal of Experimental Psychology: Learning, Memory, and Cognition. 16. 717-726. Einstein, G. O., & McDaniel, M. A. (1996). Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 115-141). Mahwah, NJ: Erlbaum. Einstein, G. O., McDaniel, M. A., Richardson, S. L., Guynn, M. J . , & Cunfer, A. R. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 996-1007. Einstein, G. O., Smith, R. E., McDaniel, M. A., & Shaw, P. (1997). Aging and prospective memory: The influence of increased demands at encoding and retrieval. Psychology and Aging, 12, 479-488. Ellis, J . A. (1988). Memory for future intentions: Investigating pulses and steps. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 1, pp. 371-376). New York, NY: John Wiley & Sons. Ellis, J . A. (1996). Prospective memory or the realization of delayed intentions: a conceptual framework for research. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective Memory: Theory and application (pp. 1-22). Mahwah, NJ: Erlbaum. 195 Ellis, J . A. (1998). Prospective memory and medicine taking. In L. B. Myers, K. Midence, et al. (Eds.), Adherence to treatment in medical conditions (pp. 113-131). Amsterdam, Netherlands: Harwood Academic Publishers. Ellis, J . A., & Kvavilashvili, L. (2000). Prospective memory 2000: Past, present, and future directions. Applied Cognitive Psychology, 14. S1-S9. Ellis, J . A., Kvavilashvili, L , & Milne, A. (1999). Experimental tests of prospective remembering: The influence of cue-event frequency on performance. British Journal of Psychology. 90, 9-23. Ellis, J . A., & Milne, A. B. (1996). Retrieval cue specificity and the realization of delayed intentions. The Quarterly Journal of Experimental Psychology, 49A, 862-887. Ellis, J . A., & Nimmo-Smith, I. (1993). Recollecting naturally occurring intentions: A study of cognitive and affective factors. Memory. 1. 107-126. Erngrund, K., Mantyla, T., & Nilsson, L.-G. (1996). Adult age differences in source memory: A population based study. Journal of Gerontology: Psychological Sciences, 51B, P335-345. Eysenck, M. W. (1981). Personality, learning and memory. In H. J . Eysenck (Ed.). A model for personality. New York, NY: Springer-Verlag. Foa, E. B., & Rothbaum, B. O. (1998). Treating the trauma of rape. New York: Guilford. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-Mental State": A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198. Foster, J . K., & Jelicic, M. (Eds.) (1999). Memory: Systems, process, or function? New York, NY: Oxford University Press. 196 Freud, S. (1901). The psyche-pathology of everyday life. London: Hogarth Press. Glass, J . N. (1998). Differential subtest scores on the Rivermead Behavioural Memory Test (RBMT) in an elderly population with diagnosis of vascular or nonvascular dementia. Applied Neuropsychology, 5, 57-64. Glisky, E. L. (1996). Prospective memory and the frontal lobes. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 249-266). Mahwah, NJ : Erlbaum. Godden, D., & Baddeley, A. D. (1975). Context-dependent memory in two natural experiments: On land and underwater. British Journal of Psychology, 66, 325-331. Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54. 493-503. Gollwitzer, P. M., & Bayer, U. (1999). Deliberate versus implemental mindsets in the control of action. In S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 403-422). New York, NY: Guilford. Gollwitzer, P. M., & Brandstaetter, V. (1997). Implementation intentions and effective goal pursuit. Journal of Personality and Social Psychology, 73, 186-199. Goschke, T., & Kuhl, J . (1993). The representation of intentions: Persisting activation in memory. Journal of Experimental Psychology: Learning. Memory, and Cognition. 19, 1211-1226. Goschke, T., & Kuhl, J . (1996). Remembering what to do: Explicit and implicit memory for intentions. In M. Brandimonte, G. 0 . , Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications (pp. 53-91). Mahwah, NJ : Lawrence Erlbaum. 197 Graf, P. (1990). Life-span changes in implicit and explicit memory. Bulletin of the Psvchonomic Society. 28, 353-358. Graf, P., & Birt, A. R. (1996). Explicit and implicit memory retrieval: Intentions and strategies. In L. Reder (Ed.), Implicit memory and metacoqnition (pp. 25-44). Hillsdale, NJ : Erlbaum. Graf, P., & Mandler, G. (1984). Activation makes words more accessible, but not necessarily more retrievable. Journal of Verbal Learning and Verbal Behaviour, 23, 553-568. Graf, P., & Schacter, D. L. (1985). Implicit and explicit memory for new associations in normal and amnesic participants. Journal of Experimental Psychology: Learning. Memory, and Cognition, 11, 501-518. Graf, P., & Schacter, D. L. (1987). Selective effects of interference on implicit and explicit memory for new associations. Journal of Experimental Psychology: Learning, memory, and Cognition, 13. 45-53. Graf, P., & Uttl, B. (1999). Prospective memory: A new focus for research. Paper submitted for publication. Graf, P., & Williams, D. (1987). Completion norms for 40 three-letter word stems. Behavior Research Methods, Instruments, & Computers, 19, 422-445. Greene, R. L. (1986). Word stems as cues in recall and completion tasks. Quarterly Journal of Experimental Psychology, 38A, 663-673. Guynn, M. J . , McDaniel, M. A., & Einstein, G. O. (1998). Prospective memory: When reminders fail. Memory & Cognition. 26, 287-298. Habib, R., Jelicic, M., & Craik, F. I. M. (1996). Are implicit memory deficits in the elderly due to differences in explicit memory processes? Aging, Neuropsychology, and Cognition. 3, 264-271. 198 Hannon, R., Adams, P., Harrington, S., Fries-Dias, C , & Gipson, M. T. (1995). Effects of brain injury and age on prospective memory self-rating and performance. Rehabilitation Psychology. 40. 289-298. Harris, J . E. (1980). Memory aids people use: Two interview studies. Memory and Cognition. 8. 31-38. Harris, J . E. (1984). Remembering to do things: A forgotten topic. In J . E. Harris, & P. E. Morris (Eds.), Everyday memory, actions, and absentmindedness (pp. 71-92). London: Academic Press. Harris, L. M., & Menzies, R. G. (1999). Mood and prospective memory. Memory. 7. 117-127. Harris, J . , & Wilkins, A. J . (1982). Remembering to do things: A theoretical framework and illustrative experiment. Human Learning. 1. 1-14. Hasher, L., & Zacks, R. T. (1979). Automatic and effortful processes in memory. Journal of Experimental Psychology: General. 108. 356-388. Hayman, C. A. G. , & Tulving, E. (1989). Is priming in fragment completion based on a "traceless" memory system? Journal of Experimental Psychology: Learning. Memory, and Cognition. 15. 941-956. Herrmann, D. J . (1984). Questionnaires about memory. In J . E. Harris,. & P. E. Morris (Eds.), Everyday memory, actions, and absentmindedness (pp. 133-151). London: Academic Press. Herrmann, D. J . , & Chaffin, R. (1988). Memory in historical perspective: The literature before Ebbinghaus. New York, NY: Springer-Verlag. Herrmann, D. J . , & Neisser, U. (1978). An inventory of everyday memory experiences. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory (pp. 52-60). New York, NY: Academic Press. 199 Hertel, P. T., & Rude, S. S. (1991). Depressive deficits in memory: Focusing attention improves subsequent recall. Journal of Experimental Psychology: General, 120, 301-109. Hertzog, C , & Hultsch, D. F. (2000). Metacognition in adulthood and old age. In F. I. M Craik, & T. A. Salthouse (Eds.), The handbook of aging and cognition (pp. 417-466). Mahwah, NJ: Lawrence Erlbaum. Hertzog, C , Park, D. C , Morrell, R. W., & Martin, M. (2000). Ask and ye shall receive: Behavioral specificity in the accuracy of subjective memory complaints. Applied Cognitive Psychology, 14, 257-275. Hintzman, D. L., & Hartry, A. L. (1990). Item effects in recognition and fragment completion: Contingency relations vary for different subsets of words. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16, 955-969. Hitch, G. J . , & Ferguson, J . (1991). Prospective memory for future intentions: Some comparisons with memory for past events. European Journal of Cognitive Psychology, 3, 285-296. Howe, M. L., Rabinowitz, R. M., & Grant, M. J . (1993). On measuring the (in)dependence of cognitive processes. Psychological Review, 100, 737-747. Hunt, R. R., & Smith, R. E. (1996). Representations may be restrictive: Where is the feeling of prospective memory? In M. Brandimonte, G. O., Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications (pp. 157-171). Mahwah, NJ: Lawrence Erlbaum. Huppert, F. A., & Beardsall, L. (1993). Prospective memory impairment as an early indicator of dementia. Journal of Clinical and Experimental Neuropsychology, 15, 805-821. 200 Huppert, F. A., Johnson, T., & Nickson, J . (2000). The high prevalence of prospective memory impairment in the elderly and in early-stage dementia: Findings from a population-based study. Applied Cognitive Psychology. 14. S63-S81. Hyman, I. E. Jr., & Billings F. J . (1998). Individual differences and the creation of false childhood memories. Memory. 6. 1-20. Jacoby, L. L. (1984). Incidental versus intentional retrieval: Remembering and awareness as separate issues. In N. Butters, & L. Squire (Eds.), The neuropsychology of memory (pp. 97-122). New York: Guilford. Jacoby, L. L. (1991). A process dissociation framework: Separating automatic from intentional uses of memory. Journal of Memory and Language. 30. 513-541. Jacoby, L. L . , & Dallas, M. (1981). On the relationship between autobiographical memory and perceptual learning. Journal of Experimental Psychology: General. 3. 306-340. Jacoby, L. L . , & Whitehouse, K. (1989). An illusion of memory: False recognition influenced by unconscious perception. Journal of Experimental Psychology: General. 118. 126-135. Jacoby, L. L . , & Witherspoon, D. (1982). Remembering without awareness. Canadian Journal of Psychology. 32. 300-324. James, W. (1890). Principles of psychology. (Vol.1). New York: Holt. James, W. (1892/1961). Psychology: The briefer course (G. Allport, Ed.). New York, NY: Harper & Brothers. Jennings, J . M., & Jacoby, L. L. (1993). Automatic and intentional uses of memory: Aging, attention, and control. Psychology and Aging. 8, 283-293. Jetter, W., Poser, U., Freeman, Jr., R. B., & Markowitsch, H. J . (1986). A verbal long term memory deficit in frontal lobe damaged patients. Cortex, 22, 229-242. 201 Johnson, M. K., & Raye, C. L. (1981). Reality monitoring. Psychological Review, 88, 67-85. Jorm, A. F. (1986). Controlled and automatic information processing in senile dementia: A review. Psychological Medicine, 16, 77-78. Kausler, D. H. (1994). Learning and memory in normal aging. Boston, MA: Academic Press. Kausler, D. H., Lichty, W., & Davis, R. T. (1985). Temporal memory for performed activities: Intentionality and adult age differences. Developmental Psychology, 21. 1132-1138. Kausler, D. H., Salthouse, T. A., & Saults, J . S. (1988). Temporal memory over the adult lifespan. American Journal of Psychology, 101. 207-215. Kerns, K. A. (2000). The CyberCruiser: An investigation of development of prospective memory in children. Journal of the International Neuropsychological Society, 6, 62-70. Kidder, D. P., Park, D. C , Hertzog, C , & Morrell, R. (1997). Prospective memory and aging: The effects of working memory and prospective memory task load. Aging, Neuropsychology, and Cognition, 4, 93-112. King, B., LeDrew, P., & Porter, G . (1982). Out and about. Thresholds 1. Ontario, Canada: Academic Press Canada. Kolers, P. A. (1975). Specificity of operations in sentence recognition. Cognitive Psychology, 7, 289-306. Kolers, P. A. (1976). Reading a year later. Journal of Experimental Psychology: Human Learning and Memory, 2, 554-565. 202 Koriat, A., Ben-Zur, H., & Nussbaum, A. (1990). Encoding information for future action: Memory for to-be-performed tasks versus memory for to-be-recalled tasks. Memory & Cognition, 18, 568-578. Koriat, A., Ben-Zur, H., & Sheffer, D. (1988). Telling the same story twice: Output monitoring and age. Journal of Memory and Language, 27, 23-39. Krishnan, H. S., & Shapiro, S. (1999). Prospective and retrospective memory for intentions: A two-component approach. Journal of Consumer Psychology, 8, 141-166. Kucera, M., & Francis, W. (1967). Computational analysis of present-day American English. Providence, R l : Brown University Press. Kuhl, J . , & Kazen, M. (1999). Volitional facilitation of difficult intentions: Joint activation of intention memory and positive affect removes Stroop interference. Journal of Experimental Psychology: General, 128, 3, 382-399. Kvavilashvili, L. (1987). Remembering intention as a distinct form of memory. British Journal of Psychology, 78, 507-518. Kvavilashvili, L. (1992). Remembering intentions: A critical review of existing experimental paradigms. Applied Cognitive Psychology, 6, 507-524. Kvavilashvili, L. (1998). Remembering intentions: Testing a new method of investigation. Applied Cognitive Psychology, 12, 533-554. Kvavilashvili, L., & Chitashvili, M. (1991). Forgetting intentions as a stable personality characteristic and the problem of its quantitative measurement. Matsne: Series in Philosophy and Psychology, 3, 67-78. Kvavilashvili, L., & Ellis, J . (1996). Varieties of intention: Some distinctions and classifications. In M. Brandimonte, G. O., Einstein, & M. A. McDaniel (Eds.), 203 Prospective memory: Theory and applications (pp. 23-51). Mahwah, NJ : Lawrence Erlbaum. La Voie, D., & Light, L. L. (1994). Adult age differences in repetition priming: A meta-analysis. Psychology and Aging, 9, 539-553. Levy, R. L., & Claravall, V. (1977). Differential effects of a phone reminder on patients with long and short between-visit intervals. Medical Care. 15, 435-438. Levy, R. L., & Clark, H. (1980). The use of an overt commitment to enhance compliance: A cautionary note. Journal of Behaviour Therapy and Experimental Psychiatry, 11. 105-107. Lewin, K. (1951). Intention, will, and need. In D. Rapaport (Ed. and Trans.), Organization of and pathology of thought. New York, NY: Columbia University Press. (Original work published in 1926). Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York, NY: Oxford University Press. Light, L. L. (1991). Memory and aging: Four hypotheses in search of data. Annual Review of Psychology, 42, 333-376. Light, L. L., & Anderson, P. A. (1983). Memory for scripts in young and older adults. Memory & Cognition, 11, 435-444. Light, L. L., & Singh, A. (1987). Implicit and explicit memory in young and older adults. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 531-541. Lovelace, E. A., & Twohig, P. T. (1990). Healthy older adults' perceptions of their memory functioning and use of mnemonics. Bulletin of the Psvchonomic Society, 28,115-118. 204. Lovibond, S. H., & Lovibond, P. F. (1995). The structure of negative emotional states; Comparison of the depression anxiety stress scales (DASS) with the Beck depression and anxiety inventories. Behaviour, Research, and Therapy, 3, 335-343. Luria, A. R. (1966). Higher cortical functions in man. New York, NY: Basic Books. Luria, A. R. (1973). The working brain: An introduction to neuropsychology (trans. B. Haigh). New York: Basic Books. Malcolm, N. A. (1977). Memory and mind. Ithaca, NY: Cornell University Press. Mandler, G. (1980). Recognizing: The judgment of previous occurrence. Psychological Review. 87, 252-271. Mandler, G. (1994). Hypermnesia, incubation, and mind popping: On remembering without really trying. In C. Umilta, & M. Moscovitch (Eds.), Attention and performance 15: Conscious and nonconscious information processing. Attention and performance series (pp. 3-33). Cambridge, MA: The MIT Press. Mantyla, T. (1993). Priming effects in prospective memory. Memory, 1. 203-218. Mantyla, T. (1994). Remembering to remember: Adult age differences in prospective memory. Journal of Gerontology: Psychological Sciences, 49, P276-P282. Mantyla, T. (1996). Activating actions and interrupting intentions: Mechanisms of retrieval sensitization in prospective memory. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 93-113). Mahwah, NJ : Erlbaum. 205 Mantyla, T., & Nilsson, L-G. (1997). Remembering to remember in adulthood: A population-based study on aging and prospective memory. Aging, Neuropsychology, and Cognition, 4, 81-92. Marsh, R. L , & Hicks, J . L. (1998). Event-based prospective memory and executive control of working memory. Journal of Experimental Psychology: Learning, Memory, and Cognition. 24, 336-349. Marsh, R. L., Hicks, J . L., & Landau, J . D. (1998). An investigation of everyday prospective memory. Memory & Cognition, 26, 633-643. Martin, M. (1986). Aging and patterns of change in everyday memory and cognition. Human Learning, 5, 63-74. Mastroianni, F., Panza, F., Solfrizzi, V., Nardo, G. A., Torres, F., Resta, F., & Capurso, A. (1996). Analysis of some aspects of learning and memorization processes in an adult population. Archives of Gerontology and Geriatrics, 5, 29-34. Mateer, C. A., Solhberg, M. M., & Crinean, J . (1987). Perceptions of memory functions in individuals with closed head injury. Journal of Head Trauma Rehabilitation. 2, 74-84. Maylor, E. A. (1990). Age and prospective memory. Quarterly Journal of Experimental Psychology, 42A, 471-493. Maylor, E. A. (1993a). Aging and forgetting in prospective and retrospective memory tasks. Psychology and Aging. 3. 420-428. Maylor, E. A. (1993b). Minimized prospective memory loss in old age. In J . Cerella, J . Rybash, W. Hoyer, & M. L. Commons (Eds.), Adult information processing: Limits on loss (pp. 529-551). New York, NY: Academic Press. Maylor, E. A. (1995). Prospective memory in normal ageing and dementia. Neurocase. 1. 285-289. 206 Maylor, E. A. (1996a). Age-related impairments in an event-based prospective memory task. Psychology and Aging, 11, 74-78. Maylor, E. A. (1996b). Does prospective memory change with age? In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 173-197). Mahwah, NJ: Erlbaum. Maylor, E. A. (1998). Changes in event-based prospective memory across adulthood. Aging, Neuropsychology, and Cognition, 5, 107-128. McDaniel, M. A. (1995). Prospective memory: Progress and processes. In D. L. Medin (Ed.), The psychology of learning and motivation (Vol. 33, pp. 191-121). San Diego, CA: Academic Press. McDaniel, M. A., & Einstein, G. O. (1992). Aging and prospective memory: Basic findings and practical applications. In T. E. Scruggs, & M. A. Mastropieri (Eds.), Advances in learning and behavioural disabilities (Vol. 8, pp. 87-105). Greenwich, CT: JAI Press. McDaniel, M. A., & Einstein, G. O. (1993). The importance of cue familiarity and cue distinctiveness in prospective memory. Memory, 1, 23-41. McDaniel, M. A., & Einstein, G. O. (2000a, July). Prospective memory retrieval and delayed execution: Multiple memory processes and age-related effects. Paper presented at the 1 s t International Conference on Prospective Memory, Hatfield, UK.. McDaniel, M. A., & Einstein, G. O. (2000b). Strategic and automatic processes in prospective memory retrieval: A multiprocess framework. Applied Cognitive Psychology, 14, S127-S144. McDaniel, M. A., Glisky, E. L , Rubin, S. R., Guynn, M. J . , & Routhieaux, B. C. (1999). Prospective memory: A neuropsychological study. Neuropsychology, 13, 103-110. 207 McDaniel, M. A., Robinson-Riegler, B., & Einstein, G. O. (1998). Prospective remembering: Perceptually driven or conceptually driven processes? Memory & Cognition. 26. 121-134. McDonald-Miszczak, L , Gould, O. N., & Tychynski, D. (1999). Metamemory predictors of prospective and retrospective memory performance. The Journal of General Psychology, 126. 37-52. McDowell I, Kristjansson B, Hill GB , & Hebert R. (1997). Community screening for dementia: The Mini Mental State Exam (MMSE) and Modified Mini Mental State (3MS) compared. Journal of Clinical Epidemiology. 50. 377-383. McGann, D., Ellis, J . , & Milne, A. (2000, July). Effect of concurrent task processes and divided attention on prospective memory. Paper presented at the 1 s t International Conference on Prospective Memory, Hatfield, UK. Meacham, J . A. (1982). A note on remembering to execute planned actions. Journal of Applied Developmental Psychology. 3. 121-133. Meacham, J . A. (1988). Interpersonal relations and prospective remembering. In M. M. Gruneberg, P. E. Morris, & R. N Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 1, pp. 354-359). Chichester, England: Wiley. Meacham, J . A., & Columbo, J . A. (1980): External retrieval cues facilitate prospective remembering in children. Journal of Educational Research. 73. 299-301. Meacham, J . A., & Kushner, S. (1980). Anxiety, prospective remembering, and performance of planned actions. Journal of General Psychology. 103. 203-209. Meacham, J . A., & Leiman, B. (1975/1982). Remembering to perform future actions. In U. Neisser (Ed.), Memory observed: Remembering in natural contexts (pp. 327-336). San Francisco, CA: W. H. Freeman & Co. 208 Meacham, J . A., & Singer, J . (1977). Incentive effects in prospective remembering. Journal of Psychology. 97. 191-197. Meier, B., & Graf, P. (2000). Transfer appropriate processing for prospective memory tests. Applied Cognitive Psychology. 14. S11-S27. Morris, C. D., Bransford, J . D., & Franks, J . J . (1977). Levels of processing versus transfer appropriate processing. Journal of Verbal Learning and Verbal Behaviour. 16. 519-533. Morris, P. E. (1992). Prospective memory: Remembering to do things. In M. Gruneberg, & P. E. Morris (Eds.), Aspects of memory (2nd ed., pp. 196-222). New York, NY: Routledge. Moscovitch, M. (1982). A neuropsychological approach to memory and perception in normal and pathological aging. In F. I. M. Craik, & S. Trehub (Eds.), Aging and cognitive processes (pp. 55-78). New York, NY: Plenum. Moscovitch, M. (1994). Memory and working with memory: Evaluation of a component process model and comparisons with other models. In D. L. Schacter, & E. Tulving (Eds.), Memory systems 94 (pp. 269-310). Cambridge, MA: MIT Press. Muris, P., & Merckelbach, H., Horselenberg, R. (1996). Individual differences in thought suppression. The White Bear Suppression Inventory: Factor structure, reliability, validity, and correlates. Behaviour Research and Therapy. 34. 501-513. Neisser, U. (1967). Cognitive psychology. New York, NY: Appleton-Century-Crofts. Nelson, H. (1982). The National Adult Reading Test. Windsor, Berkshire, England: NFER-Nelson. Norman, D. A. (1981). Categorization of action slips. Psychological Review. 88, 1-15. 209 Norman, D. A. (1988). The psychology of everyday things. New York, NY: Basic Books. Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behaviour. In R. J . Davidson, G. E. Schwartz, & D. Shapiro (Eds.), Consciousness and self-regulation: Advances in research (Vol. 4, pp. 1-18). New York, NY: Plenum. Nyberg, L , & Tulving, E. (1996). Classifying human long-term memory: Evidence from converging dissociations. European Journal of Cognitive Psychology, 8,163-183. Ohta, R. J . , & Kirasic, K. C. (1983). The investigation of environmental learning in the elderly. In G. D. Rowles, & R. J . Ohta (Eds.), Aging and milieu: Environmental perspectives on growing old (pp. 83-95). New York, NY: Academic Press. Okuda, J . , Fujii, T., Yamadori, A., Kawashima, R., Tsukiura, R., Fukatsu, R., Suzuki, K., Ito, M., & Fukuda, H. (1998). Participation of the prefrontal cortices in prospective memory: Evidence from a PET study in humans. Neuroscience Letters, 253, 127-130. Otani, J . , Landau, J . , Libkuman, T. M., St. Louis, J . P., Kazen, J . K., & Throne, G. W. (1997). Prospective memory and divided attention. Memory, 5, 343-360. Ouellette, J . A., & Wood, W. (1998). Habit and intention in everyday life: The multiple processes by which past behaviour predicts future behaviour. Psychological Bulletin. 124. 54-74. Park, D. C , Hertzog, C , Kidder, D. P., Morrell, R. W., & Mayhorn, C. B. (1997). Effect of age on event-based and time-based prospective memory. Psychology and Aging. 12. 314-327. 210 Park, D. C , Hertzog, C , Leventhal, H., Morrell, R. W., Leventhal, E., Birchmore, D., Martin, M., & Bennett, J . (1999). Medication adherence in rheumatoid arthritis patients: Older is wiser. Journal of the American Geriatrics Society, 47, 172-183. Park, D. C , & Kidder, D. P. (1996). Prospective memory and medication adherence. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 369-390). Mahwah, NJ : Erlbaum. Patton, G. W., & Meit, M. (1993). Effect of aging on prospective and incidental memory. Experimental Aging Research, 19. 165-176. Porter, S., & Birt, A. R. (2000). Is traumatic memory special? A comparison of traumatic memory characteristics with memory for other emotional life experiences. Submitted to Applied Cognitive Psychology. Porter, S., Birt, A. R., Yuille, J . C , & Lehman, D. (2000). Negotiating false memories: Interviewer and rememberer characteristics relate to memory distortion. Psychological Science, 11, 513-516. Posner, M. I. (1973). Cognition: An introduction. Glenview, IL: Scott, Foresman. Rabbitt, P. (1996). Why are studies of "prospective memory" planless? In M. Brandimonte, G . O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 239-248). Mahwah, NJ : Erlbaum. Rabbitt, P., & Abson, V. (1990). "Lost and found": Some logical and methodological limitations of self-report questionnaires as tools to study cognitive ageing. British Journal of Psychology, 81. 1-16. Raven, J . C. (1984). Manual for the Coloured Progressive Matrices (Rev.). Windsor, Berkshire, England: NFER-Nelson. 211 Reeves, M. B., & Dobbs, A. R. (1992, October). The utilization and evaluation of prospective memory aids by younger and older adults. Poster presented at the 21st Annual Scientific and Educational Meeting of the Canadian Association on Gerontology, Edmonton, Alberta. Rendell, P. G. , & Thomson, D. M. (1993). The effect of ageing on remembering to remember: An investigation of simulated medication regimens. Australian Journal on Ageing, 12, 11-18. Rendell, P. G. , & Thomson, D. M. (1999). Aging and prospective memory: Differences between naturalistic and laboratory tests. Journal of Gerontology: Psychological Sciences, 54B, P256-P269. Robinson-Riegler, M. B. (1994). The recognition-recall hypothesis of prospective memory. Unpublished doctoral dissertation, Purdue University, W. Lafayette, IN. Roediger, H. L. (1984). Does current evidence from dissociation experiments favor the episodic/semantic distinction? Behavioural and Brain Sciences, 7, 252-254. Roediger, H. L. (1990). Implicit memory: A commentary. Bulletin of the Psvchonomic Society, 28, 373-380. Roediger, H. L. (1996). Prospective memory and episodic memory. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and application (pp. 149-155). Mahwah, NJ: Erlbaum. Roediger, H. L., Buckner, R. L., & McDermott, K. B. (1999). Components of processing. In J . K. Foster, & M. Jelicic (Eds.), Memory: Systems, process, or function? (pp. 31-65). New York, NY: Oxford University Press. 212 Roediger, H. L , & McDermott, K. B. (1993). Implicit memory in normal human subjects. In F. Boiler, & J . Graham (Eds.), Handbook of Neuropsychology. Vol. 8 (pp. 63-131). Amsterdam: Elsevier. Roediger, H. L , & Srinivas, K. (1993). Specificity of operations in perceptual priming. In P. Graf, & M. E. J . Masson (Eds.), Implicit memory: New directions in cognition, development, and neuropsychology (pp. 17-48). Hillsdale, NJ : Lawrence Erlbaum. Roediger, H. L., Weldon, M. S., & Challis, B. H. (1989). Explaining dissociations between implicit and explicit measures of retention: A processing account. In H. L. Roediger, & F. I. M. Craik (Eds.), Varieties of memory and consciousness: Essays in honour of Endel Tulving (pp. 3-14). Hillsdale, NJ: Lawrence Erlbaum. Rude, S. S., Hertel, P. T., Jarrold, W., Covich, J . , & Hedlund, S. (1999). Depression-related impairments in prospective memory. Cognition and Emotion, 13, 267-276. Salthouse, T. A. (1982). Adult cognition. New York, NY: Springer-Verlag. Salthouse, T. A. (1985). A theory of cognitive aging. In G. E. Stlemach, & P. A Vroon (Eds.), Advances in psychology (Vol. 28). New York, NY: Elsevier. Salthouse, T. A. (1988). Resource-reduction interpretations of cognitive aging. Developmental Reviews. 8, 238-272. Salthouse, T. A. (1990). Working memory as a processing resource in cognitive aging. Developmental Review, 10, 101-124. Salthouse, T. A. (1996). The processing speed theory of adult age differences in cognition. Psychological Review, 103, 403-428. Salthouse, T. A., & Babcock, R. L. (1991). Decomposing adult age differences in working memory. Developmental Psychology, 27, 763-776. 213 Schacter, D. L. (1992). Understanding implicit memory: A cognitive neuroscience approach. American Psychologist, 47, 559-569. Schacter, D. L , Bowers, J . , & Booker, J . (1989). Intention, awareness, and implicit memory: The retrieval intentionality criteria. In S. Lewandowsky, J . C. Dunn, & K. Kirsner (Eds.), Implicit memory: Theoretical issues (pp. 47-65). Hillsdale, NJ : Lawrence Erlbaum. Schacter, D. L., & Graf, P. (1989). Modality specificity of implicit memory for new associations. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 3-12. Schacter, D. L., Kaszniak, A. W., Kihlstrom, J . F., & Valdiserri, M. (1991). The relation between source memory and aging. Psychology and Aging, 6, 559-568. Schacter, D. L., & Tulving, E. (Eds.). Memory systems 1994. Cambridge, MA: MIT Press. Schonfield, A. E. D., & Stones, M. J . (1979). Remembering and aging. In J . F. Kilhstrom, & F. J . Evans (Eds.), Functional disorders of memory (pp. 103-139). Hillsdale, NJ : Lawrence Erlbaum. Searle, H.R. (1983). Intentionality. Cambridge: Cambridge University Press. Searleman, A., & Gaydusek, K. A. (1989, November). Relationship between prospective memory ability and selective personality variables. Paper presented at the annual meeting of the Psychonomic Society, Atlanta. Shallice, T., & P. W. Burgess (1991). Deficits in strategy application following frontal lobe damage in man. Brain, 114, 727-741. Shapiro, S., & Krishnan, H. S. (1999). Consumer memory for intentions: A prospective memory perspective. Journal of Experimental Psychology: Applied, 5, 169-189. 214 Sherry, D. F., & Schacter, D. L. (1987). The evolution of multiple memory systems. Psychological Review, 94, 439-454. Shimamura, A. P. (1990). Forms of memory: Issues and directions. In J . L. McGaugh, N. M. Weinberger, & G. Lynch (Eds.), Brain organization and memory: Cells, systems, and circuits (pp. 159-173). New York, NY: Oxford University Press. Shimamura, A. P. (1994). Memory and frontal lobe function. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 803-813). Cambridge, MA: MIT Press. Shimamura, A. P., Janowsky, J . S., & Squire, L. R. (1991). What is the role of frontal lobe damage in memory disorders? In H. S. Levin, H. M., Eisenberg, & A. L. Benton (Eds.), Frontal lobe function and dysfunction (pp. 173-195). New York, NY: Oxford University Press. Sinnott, J . D. (1986). Prospective/intentional and incidental everyday memory: Effects of age and passage of time. Psychology and Aging, 1, 110-116. Slamecka, N. J . , & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4, 592-604. Smith, S. M., & Blankenship, S. E. (1991). Incubation and the persistence of fixation in problem solving. American Journal of Psychology, 104, 61-87. Snodgrass, J . G. (1989). How many memory systems are there really?: Some evidence from the picture fragment completion task. In C. Izawa (Ed.), Current issues in cognitive processes. The Tulane Floweree Symposium on Cognition (pp. 135-173). Hillsdale, NJ : Lawrence Erlbaum. Somerville, S. C , Wellman, H. M., & Cultice, J . C. (1983). Young children's deliberate reminding. Journal of Genetic Psychology, 143, 87-96. 215 Spitzer, R. L , Williams, J . B., & Gibbon, M. (1987). Structured Clinical Interview for DSM-III-R. New York, NY: New York Sate Psychiatric Institute. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Administration, norms, and commentary (2nd ed.). New York, NY: Oxford University Press. Squire, L. R. (1987). Memory and brain. New York, NY: Oxford University Press. Squire, L. R. (1992). Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychological Review, 99, 195-231. Squire, L. R. (1993). The organization of declarative and nondeclarative memory. In T. Ono, L. R. Squire, M. E. Raichle, D. I. Perrett, & M Fukuda (Eds.), Brain mechanisms of perception and memory: From neuron to behavior (pp. 219-227). New York, NY: Oxford University Press. Squire, L. R., Knowlton, B., & Musen, G. (1993). The structure and organization of memory. In L. W. Porter & M. R. Rosenzweig (Eds.), Annual review of psychology (Vol. 44, pp. 453-495). Palo Alto, CA: Annual Review Inc. Ste-Marie, D.M. & Jacoby, L.L. (1993). Spontaneous versus directed recognition: The relativity of automaticity. Journal of Experimental Psychology: Learning, Memory & Cognition, 19, 777-788. Stuss, D. T., & Benson, D. L. (1984). Neuropsychological studies of the frontal lobes. Psychological Bulletin, 95, 3-28. Sunderland, A., Watts, K., Baddeley, A. D., & Harris, J . E. (1986). Subjective memory assessment and test performance in elderly adults. Journal of Gerontology, 4JL 376-384. 216 Taylor, J . L , Miller, T. P., & Tinklenberg, J . R. (1992). Correlates of memory decline: 4-year longitudinal study of older adults with memory complaints. Psychology and Aging. 7. 185-193. Teng, E. L , & Chui, H. C. (1987). The Modified Mini-Mental State (3MS) examination. Journal of Clinical Psychiatry. 48. 314-318. Terry, W. S. (1988). Everyday forgetting: Data from a diary study. Psychological Reports. 62. 299-303. Thorndike, E. L , & Lorge, I. (1944). The teacher's word book of 30.000 words. New York, NY: Teachers College Press, Columbia University. Tombaugh, T. N., Grandmaison, L. J . , & Schmidt, J . P. (1995). Prospective memory: Relationship to age and retrospective memory in the Learning and Memory Battery (LAMB). The Clinical Neuropsychologist. 9. 135-142. Toth, J . P., & Hunt, R. R. (1999). Not one versus many, but zero versus any: Structure and function in the context of the multiple memory systems debate. In J . K. Foster, & M. Jelicic (Eds.), Memory: Systems, process, or function? (pp. 232-272). New York, NY: Oxford University Press. Trenerry, M. R., Crosson, B., DeBoe, J . , & Leber, W. R. (1990). Visual Search and Attention Test. Odessa, FL: Psychological Assessment Resources. Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization and memory (pp. 381-403). New York, NY: Academic Press. Tulving, E. (1983). Elements of episodic memory. New York, NY: Oxford University Press. Tulving, E. (1985a). How many memory systems are there? American Psychologist. 40. 385-398. 217 Tulving, E. (1985b). Memory and consciousness. Canadian Psychology, 26, 1 -12. Tulving, E. (1993). Varieties of consciousness and levels of awareness in memory. In. A. Baddeley, & L. Weiskrantz (Eds.), Attention, selection, awareness, and control: A tribute to Donald Broadbent (pp. 283-299). Oxford: Clarendon Press. Tulving, E. (1999a). On the uniqueness of episodic memory. In L-G. Nilsson, & H. J . Markowitsch (Eds.), Cognitive neuroscience of memory (pp. 11-42). Toronto, Canada: Hogrefe & Huber Publishers. Tulving, E. (1999b). Study of memory: Processes and systems. In J . K. Foster, & M. Jelicic (Eds.), Memory: Systems, process, or function? (pp. 11-30). New York, NY: Oxford University Press. Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science. 247, 301-306. Tulving, E., Schacter, D. L , & Stark, H., A. (1982). Priming effects in word-fragment completion are independent of recognition memory. Journal of Experimental psychology: Learning. Memory, and Cognition, 8, 336-342. Tulving, E., & Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80, 352-373. Uttl, B., & Graf, P. (1993). Episodic spatial memory in adulthood. Psychology and Aging. 8. 257-273. Uttl, B., Graf, P., Miller, J . , Mclsaac, H., & Tuokko, H. (1999). Pro- and retrospective memory in late adulthood. Manuscript submitted for publication, Cognition and Consciousness. 218 van der Hout, M., Merckelbach, H., & Pool, K. (1996). Dissociation, reality monitoring, trauma, and thought suppression. Behavioural and Cognitive Psychotherapy. 24. 97-108. Verhaeghen, P., Marcoen, A., & Goossens, L. (1993). Facts and fiction about memory aging: A quantitative integration of research findings. Journal of Gerontology: Psychological Sciences, 48. P157-P171. Watkins, M. J . , & Peynircioglu, Z. (1983). On the nature of word recall: Evidence for linguistic specificity. Journal of Verbal Learning and Verbal Behaviour. 22,385-394. Waugh, N. C , & Norman, D. A. (1965). Primary memory. Psychological Review. 72. 89-104. Webster's new compact format dictionary. (1988). Larchmont, NY: Book Essentials. Wechsler, D. (1955). Wechsler Adult Intelligence Scale (WAIS). New York, NY: Psychological Corporation. Wegner, D. M., Schneider, D. J . , Carter, S. R., & White, T. L. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology. 53. 5-13. Wegner, D. M., & Zanakos, S. (1994). Chronic thought suppression. Journal of Personality. 62. 615-640. Weldon, M. S. (1999). The memory chop shop: Issues in the search for memory systems. In J . K. Foster, & M. Jelicic (Eds.), Memory: Systems, process, or function? (pp. 162-204). New York, NY: Oxford University Press. Weldon, M. S., & Roediger, H. L. (1987). Altering retrieval demands reverses the picture superiority effect. Memory and Cognition, 15, 269-280. 219 Wenzlaff, R. M., & Wegner, D. M. (2000). Thought suppression. Annual Review of Psychology, 51, 59-91. West, R. L. (1988). Prospective memory and aging: In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 2, pp. 199-125). Chichester, England: Wiley. West, R., & Craik, F. I. M. (1999). Age-related decline in prospective memory: The roles of cue accessibility and cue sensitivity. Psychology and Aging, 14, 264-272. West, R., Herndon, R. W., & Ross-Munroe, K. (2000). Event-related neural activity associated with prospective remembering. Applied Cognitive Psychology, 14, S115-S126. Wheeler, M. A., Stuss, D. T., & Tulving, E. (1997). Toward a theory of episodic memory: The frontal lobes and autonoetic consciousness. Psychological Bulletin, 121, 331-354. Wichman, H., & Oyasato, A. (1983). Effects of locus of control and task complexity on prospective remembering. Human Factors, 25, 583-591. Wilkins, A. J . , & Baddeley, A. D. (1978). Remembering to recall in everyday life: An approach to absent-mindedness. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues: Memory in everyday life (Vol. 1, pp. 27-34). New York, NY: John Wiley & Sons. Wilson, B. A., Cockburn, J . , & Baddeley, A. D. (1985). The Rivermead Behavioural Memory Test. Reading, Berkshire, England: Thames Valley Test Company. Winograd, E. (1988). Some observations on prospective remembering. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: 220 Current research and issues: Memory in everyday life (Vol. 1, pp. 348-353). New York, NY: John Wiley & Sons. Worthington, A. (1999). Dysexecutive paramnesia: Strategic retrieval deficits in retrospective and prospective remembering. Neurocase, 5, 47-57. Zeigarnik, B. (1938). On finished and unfinished tasks. In W. D. Ellis (Ed.), A source book of Gestalt psychology (pp. 300-314). New York, NY: Harcourt, Brace, & World. (Reprinted and condensed from 1927). 221 Table 1. Summary of the Number of Dissociations Between Prospective (ProM) and Retrospective (RetM) Memory Task Performance Demonstrated in the Research to Date. 3 RetM Task Degree/Type of Dissociation Stochastic Dissociations Strong Normal None Free Recall 1 19 9 Recognition — 8 2 STM 10 1 Implicit — 1 2 Column% 1.9 71.7 26.4 (0=53) Functional Dissociations Double Single None Free Recall 3 25 20 Recognition 1 10 7 STM ~ 11 14 Implicit — — 1 Column% 4.3 50.0 45.7 (n=92) Totals n (%) Strong/Double Normal/Single None Row% Free Recall 4 (2.8)b 44 (30.3) 29 (20.0) 53.1 Recognition 1 (0.7) 18 (12.4) 9 (6.2) 19.3 STM — 21 (14.5) 15(10.3) 24.8 Implicit — 1 (0.7) 3(2.1) 2.8 Column% 3.5 57.9 38.6 100 (N=145) Note. a Three prospective-retrospective task comparisons were not included in this table under functional dissociations because for two the retrospective task was a semantic memory task and for the other it was a Wechsler Memory Scale-Revised score. b Numbers in parentheses are percentages out of total number of observations (N = 145). 222 Table 2 Participant Demographic and Health Characteristics. Measure Young Old F(1,130) Number of participants Age Gender Male Female Ethnic background Caucasian Black/ African Canadian Asian Native Canadian/ Aboriginal Other Native language English French Other Highest educational level < high school degree high school degree partial college/ training college degree graduate degree Current year in university 66 20.09 (3.35) (Range=17-39) 14(21.2%) 52 (78.8%) 86.4% 1.5% 1.5% 3.0% 7.6% 93.9% 1.5% 4.5% 100% 66 75.42 (7.32) (Range=63-92) 14 (21.2%) 52 (78.8%) 93.9% 1.5% 1.5% 3.0% 98.5% 1.5% 18.2% 16.7% 34.8% 18.2% 12.1% 2.1 (1.1) (Range=1-5) Highest occupational level unskilled semiskilled skilled semiprofessional professional # Social clubs (scale 1-4) none between 1 and 3 between 3 and 7 over 7 # hrs. outside home (scale 1-5) none 1 to 5 hours 5 to 10 hours 10 to 20 hours over 20 hours General health (scale 1=very good; 5=very poor) Health compared to others (scale 1=much better; 5 much worse) Health interferes with activities (scale 1=not at all; 3=a great deal) Eyesight (scale 1=excellent; 5=unable to see) Hearing (scale 1=excellent; 5=unable to near) 7.6% 12.1% 3.0% 22.7% 40.9% 24.2% 15.2% 40.9% 33.3% 1.88 (.71) 2.22 (.76) 7.38** 31.8% 15.2% 48.5% 51.5% 19.7% 28.8% 4.5% 4.56 (.70) 3.85(1.08) 20.01** 3.0% 1.5% 7.6% 7.6% 25.8% 24.2% 28.8% 66.7% 34.8% 1.74 (.86) 1.67 (.71) .303 2.62 (.84) 2.06 (.78) 15.81** 1.20 (.47) 1.60 (.61) 17.34** 1.65 (.67) 1.95 (.67) 6.81** 1.58 (.56) 1.94 (.72) 10.53** 224 3MS Total score (out of 100) - 95.20 (3.44) (Range=86-100) Three words (out of 3) - 2.97 (.17) First recall (out of 9) - 8.15(1.48) Second recall (out of 9) -- 8.26 (1.36) Note: Where appropriate, means (and standard deviations) are reported, otherwise frequencies (%) are reported. *rj<.05, **fj<.01, ***£<.001 225 Table 3 Main Experimental Design Age group Young Old Study Semantic Nonsemantic Semantic Nonsemantic RetM Tests Word completion Word completion Cued recall Cued recall Recognition Recognition Word completion Word completion Cued recall Cued recall Recognition Recognition ProM Tests ProM1 ProM1 ProM2 ProM1 ProM1 ProM2 Note: RetM = retrospective memory; ProM = prospective memory. Age was a between-subjects factor. Study orientation, RetM test, and ProM tests were within-subjects factors. For ProM1, only one target word was studied under each of the study orientations. The levels of processing manipulation did not apply to ProM2. 226 Table 4 Tasks Completed by all Participants Listed in Order of Administration Order Task Informed Consent 1 3MS (older participants only) 2 Study: Rating & counting task (levels of processing manipulation) 3 ProM1 instructions 4 Demographics questionnaire ^251 (ProM1 retention interval = approx. 10 minutes) 6 Implicit word completion test 3 ProM1 cues embedded within this test 7 Explicit cued recall test 3 ProM1 cues embedded within this test 8 Explicit word recognition test 3 ProM1 cues embedded within this test 9 ProM2 "Reading task" instructions 10 Modified VSAT 2 selective attention tasks (ProM1 retention interval = approx. 10 minutes) 2 divided attention tasks 11 ProM2 "Reading task" 9 ProM2 cues presented in total 12 ProM2 "Reading task" comprehension & follow-up questions 13 NAART 14 PMQ 15 DES Debriefing Note: 3MS = Modified Mini-Mental State (Teng & Chui, 1987); ProM1 = prospective memory test 1; WBSI = White Bear Suppression Inventory (Wegner & Zanakos, 1994); ProM2 = prospective memory test 2; NAART = North American Adult Reading Test (Blair & Spreen, 1989); PMQ = Prospective Memory Questionnaire (Hannon et al., 1995); DES = Dissociative Experiences Scale (Carlson & Putnam, 1993). 227 Table 5 Proportions Correct (M and SD) for the Implicit and Explicit Memory Tasks as a Function of Study Task and Age Group Young Old Task and measure Pleasantness Consonants Pleasantness Consonants Word completion Target Item-corrected scores Baseline Difference Subject-corrected scores Baseline Difference .43 (.23) .19 (.13) .23 (.25) .47 (.18) .23 (.12) .23 (.24) .23 (.10) .20 (.24) .24 (.22) .37 (.18) .24 (.16) .13 (.23) .37 (.18) .22 (.15) .15 (.24) .24 (.14) .12 (.22) .12 (.23) Cued recall Proportion correct Corrected score Word Recognition Proportion correct Corrected score .52 (.20) .35 (.22) .29 (.22) .76 (.18) .47 (.23) .56 (.16) .34 (.18) .17 (.15) .15 (.18) .60 (.26) .26 (.17) .33 (.17) 228 Table 6. Prospective Memory (ProM) Test Scores as a Function of Age Test and Measure Young Old ProM Taskl Strict scoring 3 .41 (.26) .20 (.25) Lenient/detailed scoring13 .44 (.27) .27 (.26) Word completion context0 .67 (.36) .42 (.39) Cued recall context0 .42 (.39) .32 (.37) Word recognition context0 .21 (.31) .06 (.20) Initial performance0 , 60.6% 36.4% Probability of forgetting6 .38 (.15) n=54 .38 (.14) n=45 Probability of recovery 6 .35 (.15) n=41 .39 (.14) n=32 ProM Task2 Strict scoring 3 .61 (.32) .31 (.36) Lenient/detailed scoring 0 .66 (.33) .35 (.39) Mean number of errors f 3.50 (2.89) 6.17 (3.28) No error (correct response) 5.50 (2.89) 2.83 (3.28) Late responses 9 .76 (1.01) .61 (.89) Slight errors .53 (.73) .27 (.60) Intermediate errors .23 (.58) .33 (.66) Omissions/ Serious errors 2.74 (3.06) 5.56 (3.76) Initial performance0 , 75.8% 31.8% Probability of forgetting6 .45(.20)n=38 .41 (.16)n=23 Probability of recovery 6 .37(.16)n=32 .39 (.13)n=25 ProM Questionnaire (PMQ)h 2.07 (.75) 1.34 (.74) Long-term episodic 1.61 (1.02) .67 (.73) Short-term habitual .43 (.42) .17 (.25) Internally cued 2.49 (1.35) 1.46 (1.04) Techniques to remember 3.64 (1.66) 2.91 (1.64) 229 a Mean (SD) proportion correct (out of 9). b Mean (SD) proportion (maximum score=27). 0 Mean (SD) proportion (maximum = 9). d Percentage of participants who responded to the first cue. e See text for details. f All errors are indicated in terms of mean number (SD). 9 Late responses are the sum of slight and intermediate errors. h Mean response (SD) on a 9-point Likert rating scale (anchor = 0). 230 Table 7 Mean Ratings of Task Difficulty, Interest, Absorption, and Willingness to Comply and Comprehension Scores for Prospective Memory Task 2: The Reading Task Young Old Measure (M and SD) (M and SD) Difficulty313 1.95 (.61) 1.79 (.73) n=57 n=34 Interest3 2.05 (.79) 2.61 (.87) n=66 n=66 Absorption 3 2.32 (.75) 2.82 (.82) n=66 n=66 Willingness 3 b 3.32 (.57) 3.56 (.56) n=57 n=34 Comprehension 0 3.82 (1.55) 2.65 (1.42) n=66 n=66 3 Ratings were made on four-point scales (see text for details). b Includes only those participants who made at least one response in the reading task. c Maximum score = 6. 231 Table 8 Summary of the MANOVA Assessing the Effects of Age (Two Groups) on the Retrospective and Prospective Memory Tests Memory Test Young (M and SD) Old (M and SD) Age Difference3 Effect Size d Word completion** .23 (.18) .14 (.18) .49 Cued recall*** .29 (.22) .15 (.18) .69 Word recognition*** .56 (.16) .33 (.17) 1.39 Prospective memory test 1*** .44 (.27) .27 (.26) .64 Prospective memory test 2*** .66 (.33) .35 (.39) .86 *p < .05, **p < .01, ***£ < .001 3 Effect size = Hedge's d (mean difference/ pooled SD) 232 Table 9 Summary of the MANOVA Assessing the Effects of Age (Three Groups) on the Retrospective and Prospective Memory Tests. Memory Test Young (M and SD) n = 66 Old (M and SD) n = 34 Oldest (M and SD) n = 32 Word completion** .23 (.18)a .14 (.22)ab .14 (.14)° Cued recall*** .29 (.22)a .19 (.17)° .10 (.18)° Word recognition*** .56 (.16)a .33 (.18)° .33 (.15)° Prospective memory test 1 *** .44 (.27)a .32 (.27)ab .21 (.24)° Prospective memory test 2*** .66 (.33)a .43 (.40)° .26 (.37)° *e<.05, * *e< .o i , ***p<.ooi Note. Letter superscripts (a, b, c) indicate the means that are significantly different from one another and those that are not, with a being significantly higher than b and b higher than c (p < .05). Means with the same superscripts do not differ. 233 Table 10. Effect of Study on Prospective Memory Performance (M and SD) Young Old Study variable Pleasantness Consonant Pleasantness Consonants s Study Orientation .21 (.23) .21 (.22) .15 (.19) .10 (.17) Studied Nonstudied Studied Nonstudied Study vs. No .23 (.20) .23 (.18) .14 (.17) .18 (.20) study Note: These task scores apply to Prospective Memory Test 1 only. Prospective Memory Test 2 scores are not included. 234 Table 11 Pattern of Correlations Between the Memory Tests: Partial Correlations Controlling for Age and Correlations for the Young and Old Age Groups Separately Word Cued Word ProM1 ProM2 completion recall recognition Partial Correlations, Controlling for Aqe (n = 132) ProM1 — ProM2 .376** ~ Word completion .075 .041 — Cued recall .317** .350** .205* — Word recognition .319** .136 .092 .379** Younq (n = 66) ProM1 ~ ProM2 .200 — Word completion .098 -.012 — Cued recall .203 .340** .295* — Word recognition .338** .100 .190 .530** Old (n = 66) ProM1 ~ ProM2 .553** — Word completion .061 .094 ~ Cued recall .490** .397** .013 — Word recognition .322** .193 .000 .220 Note: ProM = prospective memory test *rj<.05; **p.<.01 235 Table 12. Individual Difference Measures as a Function of Age Test Young Old (M and SD) (M and SD) WBSI 46.06 (11.05) 43.71 (12.98) DES** * 15.39 (9.78) 5.91 (4.43) Modified V S A T (total score)*** 257.33 (43.11) 171.21 (48.66) Selective attention*** 152.36 (20.41) 102.20 (29.82) Divided attention*** 104.97 (31.34) 69.02 (22.19) NAART (# correct)*** 31.44 (8.61) 38.41 (11.81) Full Scale IQ*** 104.74 (6.72) 110.07 (9.16) Verbal IQ*** 102.39 (7.66) 108.47 (10.45) Performance IQ*** 106.98 (3.62) 109.85 (4.93) *2<-05, **p_<.01, ***p_<.001 CO CO CN I Si 3 3 O CO CD O O CO CO (0 CD o CO o c! CD CO CD L -3 CO co (D c: CD CD 5 © m CO c: ^ ° CO _co JD CD - Q a o h- ol o ^ CD CN O O (SI O O C N - - H c \ l - - . - - r - r ^ . J n i D C S N ^ O l N l b J l n r - r - o o o o c o c n 1 - : • f l - C O T r o C N C N ^ T J - O C D m o CN CO co m co o o CN CN CN o o o o o o - m CN CN T o T T CN *— CN CN CN CN •<- CN — O a. o i C L ^- i - CD T T o *-_ £ « 3 g . o CO cr C CO i • I i « s S H t- f= CD - O Q . _ l ( f l £ h U J "> — Q . ~ 3 CD O C O CD O O — O b CD CD CO CO o CD CD O X X CO CD E o CD • g m O T3 CD V Ql in o v Ql * CO I CD > co CD c o 15 o T3 LU E © i o .c CO II I -E u_ B CJ) c o CO co O T - N r> OJ T -CO LU Q CD < < CO CO > > h- CO 2 o CO CM §1 o <D o ©I co o o CO (0 (0 o E (D (D > O o x> c ra en <D CO re CO • g > T3 C C <D © © m CO c (0 o I— O CO ° s co co o CO T-b § 5 fo i - <o oo o) ~ ^ ^ : co LJ o co CM CO CD o o 2 o Q . II 2 2 Q . Q . O 3 Q . to CD -£= _ l CO i CO CD o T-2 m l ) CM T 3 CO o „ t =5 5" ° 5 C CO E x: o CO O =3 o o o co co .Z o o O T o 10 CO co i I Q . CO U CD o o co o CD X CD ,0) CJ CO CJ o CO o .c CD 1 3 CO m O T - CM CD O) T - r -CD > T 3 •Q «D CD 5 O > to T 3 f— H H K < < % > > Z CO CD O —^ CM 2 o I -co < H I co co co Q CO v Ol IT) O V CD C CO (D > 15 o ro o =3 "O CO "55 to .c a) c o co o T3 UJ E £ i o to E cn c o O C D C O C M 3 O CD O ©I CO CD i _ o o CO J C CO CO © > © CO o © T3 c (0 CO © k— CO CO © CO 31 > c © © © CD co c CD B © © CO o (- O <p. * ro ro co co o CM J J CO ! CJ> cn ro _ O o o CD CM O o , — CO T CM ro CO CO to CM CM . CO CM CM CM CM CM •— CM * — CO O) o o O CM O CO ! , , CO CM CM CO CO c o r ^ m o c M i o r * - * -O O C M C M C O . - 0 0 " c O L O r - g ^ c o c n ^ ^ ^ g CO CM • ro CO T CO CM O L ? c o c o r - - c o L i c N C N ^ . ; _ - ; j — . CM O O CM J o c b r - c o g b r o ^ g ^ r o ^ C M C M ' - O . C O L O - . O ; J ; CO CD . — CO CO CO CO o CM * L; r~ CO . ro T— * — oo r- T- O cn o CD o r-o CD CD O ( n t-- cn g P ^ P o o o co o — co Q. E c o s 'E o o CJ O O cr Q. CO o 3 CO _Q Q. CO co J C "O CO CJ CD 3 cr o CD co o CO _ r-r- CM co *~ o CO o If) CO CO CO c CD CD o .c x: "5 16 o Q. CD O c CO o CD CO O o X o CM i - cz> CD £ "co CD X J ; 2 O) o o CO CO ro ro CJ o co CD E o sz CD •V CO "3 O CO CD co co LU Q ro o < CO > < co > CD < co > o V Q4 o v Qt "D CD c 3 to ~CD > _CD 16 c o CD O 3 T3 CD Do CD c o ro o 3 T3 LU E S I •c o sz CO CO E B CD o cr z o CN JCD CO CD a < o a c a o O co CD o o CO co CO (-> t o E CD CD > O CD Q 00 o c CO co CD 3 CO co CD CD O c CD i CD CO 3 T J > T J c CD CD $ a> m CO c o 15 <D o O ~co '•e co 0 . T- I f ) CO o i - •>-o o o o ; ; ; cn CD CD CN N CN CO CO m o O o m CO CO m *- o CO * I f ) CN p 2 P - o b I f ) b CN 00 • o CN O CN CN CN CN . . cb o fo i f ) o o CO o CO CO 1 CO CN CN CO CO CO CO « CN CO CO CD CO Q ay b ! o o CN O o O o CO ?M fo CO CO CN o O o ! I f ) i f ) ^ ~ o O ) CD cx> o co I f ) CN CN o CO CO CD o CN CN ' — CO 1 » ® b t § S 8 » 2 S ? g i f ) » - CD CN *— CN if) CN r~ co o o o »-00 CN 2 CO O • CO f~ o o O CO •>- o o 8 - '«> £ o P <- CM O o i i 1 £ 2 5 a. a. T 3 CD y is o O = d CD c ^ 1 o CD J9 2- - o CO Q. o o O CO CD X o CO 5 o o co CD E o _IZ CD cn o CO CO m CO CN CO CO CD > TJ - o -g < < < < co co co < > > > z CO S CO CO V ai i n q v Ql * E aj i o _c CO II co E i_ 0 i OJ c o o CN I o 2 c o O co CO o o CO CO CO CD > o <» o % c 1 CO co CD i _ CO CO CD CO "O '> T3 C c CD CD © m CO c o © o co o J D J 3 CD CO 'tr CO CM 2 CD • O • o CD O in CO o 0 CD CD CM CN C\J CO ro CO o O o * OO CO r- , . CO CM CO » -o o In 00 CO CN CO CO CD CD Q O) b • o O CN O O O o CO CM CO CO CD CM O O o ! If) If) ^ ~ O o> co o o CO If) CM CM o CO CO co r- o CM CM CO b c \ i ^ c b E ! § 0 c o CM CN CN CM . O LO CM r~ CO O O O •>-r - O) f-CN •r- O O) If) 00 CO 00 o If) o CD CM O O o o a> - S o o H _ CD O) £ o o r- If) CM O •>-C0 O Q. CO I O Z l T- CM c o O) o o CD cc CO X ) Q_ CD CD If) O T3 CD CO CM 2 CD 3 It) * $ * CD CO 1 CJ .c o CD ro CL O o O fl) £ r CD CD X o CO "5 o o co co CD E o - C CD " D CO "3 O CO m co LU Q CO < < co co > > r- co -o CO < co > o CN V QJ CO O V QJ * E B i t o .c co ll r-C0 E B C o Appendix A Summary of the patterns of stochastic and functional dissociations between prospective and retrospective memory tasks found in the prospective memory literature. A total of 44 empirical studies that included measures of both memory tasks are reviewed, yielding a total of 148 prospective-retrospective task comparisons. Study Variable ProM Task RetM Task Dissociation Task Cue Type Stochastic Dissociations Meacham & Leiman Tasks NF EB Free recall Normal (1975/1982) Wilkins & Baddeley (1978) Tasks NF TB Free recall Strong Kvavilashvili (1987) Tasks NL EB Free recall Normal (content) Einstein & McDaniel Tasks AL EB Free recall Normal (1990, Exp. 1) AL EB Recognition Normal AL EB STM Normal Einstein & McDaniel Tasks AL EB Free recall Normal (1990, Exp. 2) AL EB Recognition Normal AL EB STM Normal Maylor (1990) Tasks NF EB Free recall Normal Einstein et al. (1992, Exp. 1) Tasks AL EB Free recall None McDaniel & Einstein Tasks AL EB Recognition Normal (1993, Exp. 1) AL EB Implicit (WFC) None NL EB Recognition Normal NL EB Implicit (WFC) None McDaniel & Einstein Tasks AL EB Free recall Normal (1993, Exp. 2) AL EB Recognition Normal 243 Brandimonte & Passolunghi Tasks AL EB STM Normal (1994, Exp. 1) Brandimonte & Passolunghi Tasks AL EB STM Normal (1994, Exp. 2) Brandimonte & Passolunghi Tasks AL EB STM Normal (1994, Exp. 5) d'Ydewalle (1995) Tasks AL TB Free recall Normal Tombaugh et al. (1995) Tasks NL EB Free recall None NL EB Complex Figure None NL EB Digits forward Normal NL EB Digits backward Normal Cockburn (1996) Tasks NL EB Story recall None (1 brain injured P) NL TB Story recall Normal Burgess & Taylor (cited in Burgess & Shallice, 1997) Brain damage AL EB Free recall (instructions) None Controls AL EB Free recall (instructions) Normal Einstein etal. (1997, Exp. 1) Tasks AL EB Free recall None Kidder et al. (1997) Tasks AL EB STM Normal NL EB STM Normal Otani et al. (1997, Exp. 1) Tasks AL EB Free recall Normal AL EB Implicit (WSC) Normal Otani et al. (1997, Exp. 2) Tasks AL EB Free recall Normal Otani et al. (1997, Exp. 3) Tasks AL EB Free recall Normal Kvavilashvili (1998, Exp. 2) Tasks NL EB Story recall Normal/Str g McDaniel et al. (1998, Exp. 1) (Study/ test match) Same AL EB Free recall Normal Recognition Normal 244 Different AL EB Free recall Recognition Normal None Modulated AL EB Free recall Recognition None Normal McDaniel et al. (1998, Tasks AL EB Free recall Normal Exp. 2) AL EB Recognition Normal/Stron g Cherry & LeCompte (1999) Tasks AL EB Free recall Normal AL EB Recognition None AL EB STM Normal AL EB WM None Harris & Menzies (1999) Tasks AL EB Free recall None McDonald-Miszczak et al. (1999) Tasks NL NL EB TB Free recall (instructions) Free recall (instructions) None Normal Uttl et al. (1999) Tasks NL EB Free recall Normal Functional Dissociations Einstein & McDaniel (1990; Memory aid AL EB Free recall Normal Single Exp. 1) AL EB Recognition Normal Single AL EB STM Normal Single Age AL EB Free recall Normal Single AL EB Recognition Normal Single AL EB STM Normal Single Einstein & McDaniel (1990; Cue Familiarity AL EB Free recall Normal Single Exp. 2) AL EB Recognition Normal Single AL EB STM Normal Single Age AL EB Free recall Normal Single AL EB Recognition Normal Single AL EB STM None Shallice & Burgess (1991) Tasks (3 frontal Ps) AL/ NF AL/ NF EB/ TB EB/ TB Recall Recognition Normal Single Normal Single 245 Einstein et al. (1992, Exp. 1) Retention AL EB Free recall Normal Single Interval AL EB Recognition None AL EB STM None Einstein et al. (1992, Exp. 2) Age AL EB Free recall None AL EB Recognition None AL EB STM None Huppert & Beardsall (1993) Minimal NL EB Route recall Normal Single dementia/ dementia Mantyla (1993) Age AL EB Cued recall None Maylor (1993a) Age AL EB Free recall (I) None AL EB Free recall (D) None AL EB Picture recog. None McDaniel & Einstein (1993, Cue Familiarity AL EB Free recall None Exp. 2) AL EB Recognition None Cue AL EB Free recall None Distinctness AL EB Recognition None Rendell & Thompson (1993) Age NF EB Free recall Strong Single de Wall et al. (1994) Age NL EB Story recall (I) Normal Single NL EB Story recall (D) Normal Single NL EB Route recall (I) None NL EB Route recall (D) None Mantyla (1994) Age AL EB Cued recall None d'Ydewalle (1995) Task load x AL TB Free recall Normal Age Double Einstein et al. (1995, Exp. 2) Instruction AL EB Free recall Weak single Specificity (content) Age AL EB Free recall None (content) Einstein et al. (1997, Exp. 1) Task load AL EB Free recall Normal Single Age AL EB Free recall None 246 Einstein et al. (1997, Exp. 2) Task load x Age AL EB Free recall Normal Double Mantyla & Nilsson (1997) Age NL EB Free recall (content) Normal Single Otani et al. (1997, Exp. 1) Task load AL EB Free recall None AL EB STM Normal Single AL EB Implicit (WSC) None Otani etal. (1997, Exp. 2) Target type AL EB Free recall Normal Single AL EB STM Normal Single Otani et al. (1997, Exp. 3) # diff. targets AL EB Free recall Normal Single AL EB STM Normal Single Glass (1998) Vascular vs. Nonvascular Dementia (belonging) NL EB Story recall (I) Normal Single NL EB Story recall (D) None NL EB Route recall (I) Normal Single NL EB Route recall (D) Normal Single (appoint.) NL EB Story recall (I) None NL EB Story recall (D) Normal Single NL EB Route recall (I) None NL EB Route recall (D) None (message) NL EB Story recall (I) None NL EB Story recall (D) Normal Single NL EB Route recall (I) None - NL EB Route recall (D) None Guynn et al. (1998, Exp. 2) Reminders AL EB Free recall Recognition Normal Single Normal Single Marsh & Hicks (1998, Exp. 1) Task load AL EB STM None Marsh & Hicks (1998, Exp. Task load AL EB STM None 2) Marsh & Hicks (1998, Exp. Task load AL EB STM Normal Single 247 3) Marsh & Hicks (1998, Exp. Task load AL EB STM Normal Single 4) Marsh & Hicks (1998, Exp. Task load AL EB STM None 5) Marsh et al. (1998, Exp. 1) Recorders NF EB/ TB Free recall Normal Single Costermans & Desmette Age NL TB Free recall Normal Single (1999) d'Ydewalle et al. (1999) Age AL EB STM None AL TB STM None Krishnan & Shapiro (1999) Importance AL AL AL EB EB EB Free recall (content) Free recall (cues) Recognition (cues) None None Normal Single McDaniel et al. (1999) Frontal x Hippocampal AL EB Recognition Normal Double Frontal AL EB Free recall Normal Single AL EB Gen. knowledge None Hippocampal AL EB Free recall None AL EB Gen. knowledge Normal Single Rendell & Thompson (1999, Age NL EB Free recall None Exp. 3) NL EB Recognition None AL TB Free recall None AL TB Recognition None Rude et al. (1999) Depression AL TB WMS-R Normal Single Shapiro & Krishnan (1999, Exp. 1) Retrieval process x cue-content AL EB Free recall (content) Normal Double relatedness 248 Retrieval process Shapiro & Krishnan (1999 Content-related ness Exp. 2) Cue-relatedness Worthington (1999) Tasks (1 frontal P) AL EB Free recall Normal Single (cues) AL EB Free recall None (cues) AL EB Free recall None (content) AL EB Recognition Normal Single (cues) AL EB Free recall Normal Single (content) AL EB Recognition Normal Single (cues) NL/ EB/ Free recall Normal Single NF TB NL/ EB/ Recognition Normal Single NF TB NL/ EB/ Autobiographical None NF TB recall Note: NF = naturalistic field task; NL = naturalistic laboratory task; AL = artificial laboratory task; EB = event-based cue; TB = time-based cue; STM = short-term memory task; WM = working memory task; W F C = word fragment completion; W S C = word stem completion; P = patient; (I) = immediate; (D) = delayed; WMS-R = Wechsler Memory Scale-Revised. 249 Appendix B A list of the target words and the corresponding three-letter word stems or word fragments included in the word completion, cued recall, and word recognition tests. Shown next to each word is its number of letters, its frequency of occurrence as a first completion (as a percentage out of a maximum of 100) from Graf and Williams (G&W; 1987), and its frequency of occurrence in the norms of Thorndike and Lorge (T&L;1944) and Kucera and Francis (K&F; 1967). Target Stem Fragment # letters G&W T&L K&F abode abo_e 5 6 16 4 accuse acc e 6 2 26 10 affection aff_c on 9 2 37 18 banquet ban 7 1 18 6 brick bri 5 5 49 18 chapel cha 6 2 25 20 classic cla_s 7 1 15 36 crash era 5 ~ 34 20 decent dec t 6 3 18 20 defeat def t 6 1 50 31 deputy dep y 6 2 13 17 drain dra_n 5 2 41 18 export exp_r_ 6 3 24 10 filling fil 7 5 50 36 flood flo 5 2 50 19 formal for I 6 2 22 48 frantic fra 7 3 12 11 250 garland genuine gen grateful gra gar_a_ grove handful harvest import lantern misery mortar particle perfume planet prospect relieve rescue gro han Ian leap lea marvelous mis par per pro rel residence res restless retail salary scandal scout sal sea SCO har_e_ _ imp t mar e us mor a pla_e_ res e res_le_ ret i 7 7 8 5 7 7 6 7 4 9 6 6 8 7 6 8 7 6 9 8 6 6 7 5 3 2 3 1 3 2 3 2 3 0 3 0 1 3 4 2 11 19 37 31 17 48 31 19 50 33 26 8 17 21 34 45 46 30 25 23 9 50 13 37 9 34 25 14 13 12 17 13 14 11 15 11 21 10 21 25 13 15 29 13 20 43 8 8 251 shallow sha 7 3 27 14 shoot sho_t 5 2 50 27 spindle spi l_ 7 3 9 8 tend ten 4 1 50 43 thief thi 5 2 28 8 threat thr t 6 2 17 42 tragic tra_i_ 6 2 16 33 triumph tri_m 7 3 41 22 weapon wea n 6 2 42 42 whisper whi e_ 7 — 50 12 Note: Each target was presented as either a word stem or a word fragment at test and this was consistent across all counterbalancing conditions. A dash (--) indicates that normative data were not available for that item 253 Brain Power: An Investigation of Memory and Cognitive Abilities Description of the Research Introduction You are invited to take part in a research study in the Department of Psychology at Dalhousie University. Taking part in this study is voluntary and you may withdraw from it at any time. The study is described below. This description tells you about any potential risks, inconvenience, or discomforts that you might experience. Participating in the study may not be of any direct benefit to you, but we might learn things that will benefit others. You should discuss any questions you have about this study with the person who will explain it to you. Purpose of this Study The purpose of this study is to find out how a number of cognitive abilities, such as paying attention, generating words, reading skills, and remembering are related to one another. For example, if an individual has difficulty with concentration and attention, then it is likely that he or she will also have difficulty in performing tasks that require concentration and demand attention to complete. Therefore, the goal of this study is to measure performance on a variety of cognitive tasks, each tapping into different mental abilities, and to compare the results across these different tasks. Such a comparison will help identify which of the cognitive tasks administered are similar to one another and which are different, and will aid in increasing our understanding of memory and cognition in general. Study Design This study is a within-subjects experimental design. A number of different cognitive tests will be administered to all subjects and performance on these tests will be compared across subjects. Who Can Participate? If you are an undergraduate student at Dalhousie University, you can participate in this study. Other than that, there are no specific criteria for inclusion in this study. Who Will Be Conducting The Research? The principal investigator of this research is Angela R. Birt, a Cognitive Psychology Ph.D. student visiting Dalhousie University from the University of British Columbia. Dr. Stephen Porter, Assistant Professor of Psychology, is supervising Ms. Birt's research while she is at Dalhousie. Angela will be conducting the bulk of the research, however, several well trained research assistants will also be helping with the data collection. What Will You Be Asked To Do? If you decide to participate, you will be asked to carry out a series of simple tasks. The first task is a Rating & Counting Task in which you will be shown a series of 254 words, one at a time, and asked to either rate the pleasantness or count the number of consonants in each word. The second task is a Word Completion Task. You will be presented with a number of word puzzles (e.g., aft , aft no_n) and your job will be to write down the first word that comes to mind that fits with the letters provided. The third and fourth tasks, the Cued Recall and Word Recognition Tasks will require you to learn a short list of words and test your memory for them. The fifth task is a test of perceptual fluency. You'll be asked to search through some letters and symbols to find a specific target. The sixth task is a Reading Task designed to test reading fluency. For this task you will read a funny short story aloud. Your permission is requested to allow your reading to be audiotaped, which will be used later in scoring your performance. The seventh task is a word pronunciation task. You will be shown a list of different English words and will be asked to pronounce each one to the best of your ability. You will also be asked to fill out a few questionnaires. In many ways most of these tasks are much like the word puzzles and activities that are often printed in magazines and newspapers, and for this reason, they are only as challenging or as much fun as you choose to make them. It should take only about an hour of your time to complete these tasks. Possible Risks And Discomforts There are no risks involved with participation in this study. No physical or psychological distress is anticipated with participating in this research. Possible Benefits There are no known direct personal benefits associated with participating in this research. However, it is possible that as a result of taking part in this study, you may experience an increased awareness and knowledge of your memory and cognitive processes. Your participation in this study will help contribute to the understanding of how different memory and cognitive processes in humans work. Compensation In return for your participation, you will be awarded one (1) credit point toward your Psychology course grade. However, in accord with the guidelines of the Dalhousie University Department of Psychology Human Subject Pool, only Introductory Psychology students can be offered the credit point. Confidentiality The results from this study will be kept strictly confidential, and they will be securely stored in the locked laboratory of Dr. Stephen Porter. To ensure that your identity will be confidential, your name will not appear anywhere on the testing materials. Instead, we will assign you a randomly selected 3 digit number, only this number will appear in our records, and this number cannot be used to trace you in any manner. 256 "/ have read the explanation about this study. I have been given the opportunity to discuss it and my questions have been answered to my satisfaction. I hereby consent to take part in this study. However, I realize that my participation is voluntary and that I am free to withdraw from the study at any time. Should I decide to withdraw, I will inform the experimenter of my decision." Print Name: Date: Sign Name: Consent to be Audiotaped During the Reading Task One of the tasks in this experiment, the Reading Task, requires that participants read aloud a short story (i.e., 841 words). In order to score performance on this task accurately, we request your permission to audiotape your narration. If you agree to be audiotaped while reading this story, please sign your name and write today's date below to indicate your consent. Sign Name: Date: Appendix D An Investigation of Memory and Cognitive Abilities Response Booklet - Experimenter's Instruction Version Experimenters: Angela R. Birt, Ph.D. Candidate Dr. Kenneth Rockwood, Geriatric Medicine Research Unit, Camphill Dr. Peter Graf, Professor, University of British Columbia Dr. Stephen Porter, Assistant Professor, Dalhousie University Ethics Code Number: 2000-23 Ethics Approval Date: March 24, 2000 F00.10.SP Sept. 28, 2000 Experimenter/Tester: Participant ID#: Participant Age: Participant Gender: M F 258 Rating & Counting Task Instructions: (Read aloud and ask participants to follow along in their booklet). This task is a measure of perceptual and verbal fluency. You will be presented with 26 different words printed on index cards, one at a time. Your job will be to say each word aloud as it is presented. In addition, before each word is shown, you will be instructed to do either one of two tasks: (1) you will be asked to rate the pleasantness of the word on a scale of 1 to 5 (shown below—point out scale to participant). This is a subjective rating of pleasantness; just simply look at the word carefully as it is presented, think about its meaning, and decide how pleasant or unpleasant you feel the meaning of the word is to you. Please think about each word and rate it in such a way that you would be able to give a very similar rating if you saw the word again; or (2) you will be asked to count the number of consonants in the word; that is, count the total number of letters in the word excluding the vowels (vowels = a, e, i, o, u; y = consonant). Space is provided below for you to write your answers down as you make your responses. You will be told which task to do every time a word is presented. Before we begin, let's do a few practice trials. Rating Scale: 1 2 3 4 5 Very Somewhat Neither Pleasant Somewhat Very Unpleasant Unpleasant nor Unpleasant Pleasant Pleasant Practice Trials: (Emphasize that participants must sav the word aloud FIRST, then make their response) a (NOTE TO EXPERIMENTER: The rate of presentation of the words on the • • - cue cards is subject-paced. However participants should run through this D task fairly quickly—a couple of seconds per word). Test Responses: 1 10 19 2 11 20 3 12 21 4 13 22 5 14 23 6 15 24 7 16 25 8 17 26 9 18 259 ProM Instructions I (Read the following instructions to the participant—do not show participant this page) For the next several tasks, most of the time I will give you instructions on what I want you to do immediately before each task is to be completed. However, some of the time you will be expected to remember things on your own. For example, a little later there will be a number of different tasks in which you will be shown some word puzzles. Some of them will be word fragments with missing letters like this one: doc_0_ (show participant this example, which is on the next page) Your job will be to think of a word that successfully completes the fragment. Writing down the word in your test booklet. Can you think of a word that fits this fragment? The other word puzzle will be three letter combinations or word beginnings like this one: bla (show participant this example, which is two pages after this one) Your job will be to think of a word beginning with each set of three letters and to write each word in your test booklet. Can you think of a word that starts with the three letters "bla"? You will also be asked to do other things with these word puzzles, but I will not tell you about that right now. For now, I want you to know that among the letter combinations I will show you will be the letters "res." Every time you see the three letters "res" at the beginning of one of these word puzzles, I want you to complete the puzzle according to the specific task instructions AND spell out the word aloud. Spell out the word only when it starts with the letters "res." If you can't think of a word beginning with "res", then simply spell out the letters "r-e-s." OK? In addition, a little later you will also be shown a sheet of paper with several columns of different words on it. It will look something like this: (show participant the sample list of words, which is three pages after this one) You will be asked to do an activity with the words listed on a sheet like this, but I will not tell you about that yet. For the present purposes I want you to know that some of the words listed in the columns will begin with the letters "res." Again, every time you see a word beginning with the letters "res", I want you to spell the word out loud. Do not simply say the word aloud, instead, in every case spell it out by saying each letter of the word aloud. Now can you tell me in your own words what are you to do when you see the letters "res" either at the beginning of a word within a list or at the beginning of a word puzzle? (repeat instructions if necessary) OK. To summarize, from now on, every time you see the letters "res" at the beginning of a word or at the beginning of a word puzzle, you know have to make a response. ]t is up to you to remember these instructions. I will not remind you about this or repeat the instructions. 260 doc o 261 bla Word Recognition Task 1. senator 26. latterly 51. bitmap 2. compare 27. torpedo 52. magnolia 3. demand 28. biting 53. notice 4. bumble 29. digital 54. cruise 5. cable 30. mental 55. bloom 6. block 31. talcum 56. pedicure 7. facade 32. secrete 57. menace 8. cellar 33. bumpkin 58. facial 9. headset 34. Celtic 59. tornado 10. latchkey 35. immunize 60. factor 11. celery 36. torment 61. glazing 12. notary 37. latitude 62. demean 13. crutch 38. bloat 63. pedigree 14. glaring 39. immortal 64. mentor 15. notion 40. ridged 65. seclude 16. crunch 41. sentient 66. immobile 17.cabana 42. section 67. digest 18. crying 43. compose 68. routine 19. pedestal 44. glamour 69. demise 20. bitter 45. sensual 70. riddle 21. riding 46. crystal 71. dignity 22. cabin 47. magnet 72. hearing 23. roughly 48. healthy 73. magic 24. talker 49. talent 74. rounder 25. cryptic 50. complex 75. bumper 264 Demographics and Health Questionnaire Please answer the following questions: (Read each question aloud to the participant and either have them write their own answers or write their answers for them). 1. What is your current age? 2. What is your date of birth? (day/month/year) 3. What is your ethnic background? a. Caucasian b. Black or African Canadian c. Asian d. Native Canadian or Aboriginal e. Other (please specify) 4. What is your native language? a. English b. French c. Other (please specify) _ 5. How many years of education have you attained? 1 2 3 4 5 6 7 Less than 7th to 9th 10th to High partial college graduate 7th grade grade 11th grade school college or degree degree degree specialized training 6. What is the highest occupational level you have achieved? (If you are a student, then please rate the highest occupational level of your same-sex parent). 1 2 3 4 5 Unskilled Semiskilled Skilled Semiprofessional Professional 7. How many clubs and/or social organizations do you currently participate in? 1 2 3 4 none between 1 and 3 between 3 and 7 over 7 265 8. In general, how many hours per week do you usually spend outside your home? 1 2 3 4 5 none 1 to 5 hours 5 to 10 hours 10 to 20 hours over 20 hours 9. In general, how would you say your health is these days? 1 2 3 4 5 very good pretty good not too good poor very poor 10. How would you best describe your health compared to other people your age? 1 2 3 4 5 much better better Average worse much worse 11. To what extent does your health prevent you from participating in activities? 1 2 3 not at all somewhat a great deal 12. How good is your eyesight (with glasses, if you wear glasses)? 1 2 3 4 5 Excellent Good Somewhat Poor Unable to see Poor 13. How good is your hearing (with a hearing aid, if you wear one)? 1 2 3 4 5 Excellent Good Somewhat Poor Unable to hear Poor 266 WBSI Muris, Merckelbach, & Horselenberg (1996) (Read these instructions aloud. Have participants fill out the questionnaire at their own pace. Help them fill it out, if needed). Please indicate on a 5-point scale the extent to which you agree with the following statements. Circle the answer that most applies to you: 1. There are things I prefer not to think about. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 2. Sometimes I wonder why I have the thoughts I do. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 3. I have thoughts that I cannot stop. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 4. There are images that come to mind that I cannot erase. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 5. My thoughts frequently return to one idea. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 6. I wish I could stop thinking of certain things. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 267 7. Sometimes my mind races so fast I wish I could stop it. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 8. I always try to put problems out of mind. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 9. There are thoughts that keep jumping into my head. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 10. Sometimes I stay busy just to keep thoughts from intruding on my mind. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 11. There are things that I try not to think about. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 12. Sometimes I really wish I could stop thinking. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 13.1 often do things to distract myself from my thoughts. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 268 14.1 often have thoughts that I try to avoid. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 15. There are many thoughts that I have that I don't tell anyone. 1 2 3 4 5 Strongly Somewhat Neither agree Somewhat Strongly disagree disagree nor disagree agree agree 269 Scoring Sheet for Word Completion, Cued Recall, and Word Recognition Tests Participant ID# Scoring System: (NOTE: write all details about the participant's response in the space # points Response 3 participant remembers to spell the appropriate word out loud in response to the presentation of the target cue ("res") 2 participant responds to the target cue in way that is different from what was instructed (e.g., simply saying the word aloud, or indicating that something should be done); 1 participant responds to the wrong cue (e.g., "rel"), or responds late (after the cue is no longer shown or they skipped the cue and then went back to make a response). 0 participant fails to respond at all; does not response to the cue NOTE: Other relevant responses = indicate if (and W H E N ) participant asked about "res" instructions (e.g., "am I supposed to do the "res" thing now?"); indicate W H I C H inappropriate cues (e.g., "rel") to which responses were made; were any comments made about speci f ic responses? , etc., etc. Word Completion: ltem# Score Description/Comments about response 7 14 23 other re evant responses Cued Recall: ltem# Score Description/Comments about response 7 14 23 other re evant responses Word Recognition: ltem# Score Description/Comments about response 20 40 60 other re evant responses 270 Word Completion Task Instructions: (Read aloud and ask participants to follow along in their response booklet). This task is a measure of word fluency. It tests how quickly and efficiently you can generate words. I'm going to show you a number of word fragments with letters missing and then some three letter word beginnings, one at a time, from this binder. Your job is to look at the fragments and word beginnings and, as quickly as possible, write down the first word that comes to mind that successfully completes the word puzzle. Any English word is acceptable except for proper nouns. (Proper nouns are specific names of people, places, or things that must be capitalized). Please DO NOT use PROPER NOUNS for any of your completions. Also, please DO NOT REPEAT any of your responses! Practice: (Practice words are on the next two pages, show them one at a time, write down participants' answers in their test booklet for them or have them write their own answers). a NOTE TO EXPERIMENTER: The presentation of the fragments and stems is subject-paced. However, participants should be encouraged D to work as quickly as possible. After a full minute with no response, go on to the next cue. Further Instructions: I must warn you that some of these word puzzles may be a little difficult. If you can't think of a word that completes the puzzle, think about it a little longer. If you still cannot think of a word, then just simply say "PASS" and we will leave the space blank and move on to the next one. However, I would like you to complete as many of the puzzles as possible! Test Responses: 1 16 2 17 3 18 4 19 5 20 6 21 7 22 8 9 10 11 12 13 14 23 24 25 26 27 28 29 15 30 271 pum_ ki 272 int 273 Cued Recall Test Instructions: (Read aloud and ask participants to follow along in their response booklet). This test is designed to test your memory. For this task you are asked to think back to the words that were presented on the cue cards in the Rating & Counting task-the words you were asked to rate for pleasantness and count consonants. Similar to the previous task, I'm going to show you a number of word puzzles, one at a time, from the binder. These word puzzles are clues to help you remember the words you saw in the Rating & Counting task. So, look at each clue and try to remember a word from the cue cards that completes each puzzle. Not all of the word puzzles that I will show you relate to the previously presented words, so think carefully. Please complete them only with words that you actually remember as presented on the cue cards in the Rating & Counting task and do not make any guesses! If you write down an answer, you should be pretty certain that you saw the word before. Otherwise, just leave the space blank. Further Instructions: Again, if you can't think of a word from the cue cards that completes the word puzzle, then just say "pass", we'll leave the space blank, and move on to the next word. Remember, some of the puzzles have NOTHING to do with the words that were presented on the cue cards, so be careful! (NOTE TO EXPERIMENTER: This task is subject-paced. Recall of as many words as possible should be encouraged—Do not proceed onto the next task until the participant has not written a word for one full minute or indicates that he/she cannot remember any more words). Test Responses: (Have participants write their own answers in their booklet, or write their answers for them). 1 16 2 17 3 18 4 19 5 20 6 21 7 22 8 23 9 24 10 25 11 26 12 27 13 28 14 29 15 30 274 Word Recognition Task Instructions: (Read aloud and have participants follow along in their response booklet). This test is also designed to test your memory, but in a different way. For this task you are asked to think back again to the words that were presented on the cue cards in the Rating & Counting task-the words you were asked to rate for pleasantness and count consonants. I'm going to show you a list of 75 words printed all on one page in the binder. Only some of these words were actually presented in the Rating & Counting task, whereas others were not. So, please read each word carefully and decide whether you recognize it from the Rating & Counting task. If you do recognize the word, please write it down below beside its corresponding number. If you do not recognize the word, then just leave the space blank and go on to the next word. Be sure to write down only those words you actually remember as presented in the Rating & Counting task and do not make any guesses! (Show participant the word list) Further Instructions: Please start here at Number 1 (point to the first word), work your way down the column, come back up to the top, work your way down, and again up to the top and down the column. In other words, work with the words in order. (NOTE TO EXPERIMENTER: This task is subject-paced. Let the participant work at this task until no response has been made for one full minute or until he/she indicates that he/she is finished). 1. 26. 51. 2. 27. 52. 3. 28. 53. 4. 29. 54. 5. 30. 55. 6. 31. 56. 7. 32. 57. 8. 33. 58. 9. 34. 59. 10. 35. _ 60. 11. - 36. ' 61. 12. 37. 62. 13. 38. 63. 14. 39. 64. 15. 40. 65. 16. 41. 66. 17. 42. 67. 18. 43. 68. 19. 44. 69. 20. 45. 70. 21. 46. 71. 22. 47. 72. 23. 48. 73 24. 49. 74 25. 50. 75 275 ProM Instructions II (Read these instructions aloud to participants. Do not let them see this page). In a few minutes I'm going to get you to do a task called the "Reading Task." For this task, I will ask you to read aloud a short story called "Agnes." While reading this story, every time you encounter a word starting with the letters "res" throughout the story, instead of simply reading the word, I would like you to spell it out. For example, if while reading the story you came across the word "resume," which begins with the letters "res," you would not pronounce the word but would spell it out "r-e-s-u-m-e" instead. For your information, there may be more than one word beginning with the letters "res" in the story, or there may not be. Now, can you tell me in your own words what you have to do for the Reading Task? (repeat instructions if necessary) OK, before I ask you to read the story, I'm going to get you to do a visual attention task. Visual Search and Attention Task (Administer the VSAT now. Read primary test instructions on the test booklet aloud with participant. Remember to use a stopwatch to time each trial for 60 sec). Practice: Target = T (Do not time this trial. Just have the participant try the task, putting a slash (/) through all the targets. Instruct participants that if they make a mistake, just circle it and keep going). Further Instructions: I am going to get you to do 4 trials. The target is located at the center of the top of each page. It will be different for each trial, so look at it very carefully and cross out only those items that match the target exactly. OK. I'm going to give you 60 seconds for each trial. When I say go, please begin. Trial 1: (Target = C ) Instructions: (Instruct participant to turn the page). Look at the target carefully ... ready ... and begin! (Start timing as soon as you say begin). Trial 2: (Target = [) Instructions: This time I'm going to make the task a little more challenging. In addition to crossing out the target, I would like you to count backwards out loud from 75 at the SAME time. You will probably find this a bit more difficult to do, so just try your best. (Instruct participant to turn the page). OK, look at the target carefully ... ready ... and begin! Further Instructions: The next two trials involve colour. So, please look carefully at the target and cross out only items that match the target EXACTLY. Trial 3: (Target = Blue H ) Instructions: This time I want you cross out targets and at the SAME time count backwards out loud starting at 94. (Instruct participant to turn the page). OK, look at the target carefully....and begin! Trial 4: (Target = Blue / ) Instructions: Good. This time I'm going to give you a break. For this trial you simply have to cross out all the targets. You do not have to count aloud. (Instruct participant to turn the page). OK, look at the target carefully....and begin! Modified Visual Search and Attention Test Adapted from: Trenerry, Crosson, DeBose, & Leber (1990) 276 Participant ID# Instructions: This test is a measure of visual search. You will be presented with an array of letters or symbols and your job will be to find a particular letter or symbol embedded within the array. For example, if the target letter were T, then you would have to cross out all the Ts in each row. You are to work as quickly as you can. You will be timed and will be given a total of 60 seconds to find and cross out as many targets as you can. Example: T F I L N E V f Z K L W X I Z W L K F ^ F E T f X E F L X K I T E Z W K X N F T L I K X L W I E F X K T L E N Z F E I W X T L X E K I T L X E I F Z T W X K N F E L K I E I W I K T F L K W X Z W E F T I L X I E N T E X I E K W L X I L K F E W X Z N I E T K L F E 277 C B E F G C H E H F E C G I C H E I C B D C A C H F C B I D A C D A F C I H C I H C A H E F A C D C F E H C B F C A D E H A C E I E G C B D E G H B C A G C I E C F H I C D B C G F D C B A C B C A F C B E H F A E F E G C H C D E C B A C G D A H E B A E D G C D A F C B I F C E A D C B E A C G F E B A F E C A C F C H B C D C D G A C H E F B C A F I A B F C H D E F C G A C B E D C F A C H E H C D I C H E I C A H C E B C A F C B E H B F C A D C B D C H F B C D A G C I E C I E D H F A I C D E F B C E D C B C A G E A D F E B C G E D A C H C E D C A B A C H D C F C A C B E C F A C D G A C H E F B C A F E A B F C H D E F C G A C B E D C F A H H E G F C D B H C F C A E F A C B F C H I C F A D C A C D E F H C A D B C I F E B F E D C H A C B E C B H C I C D F E H E C A F C D E B C A G C F E D H C F B A [ / ) < ( [ ] ) ] < ) [ ( : [ ] ) : [ A [ > [ ] < [ / : \ > [ : > < [ : ] [ : ] [ > ] ) < > [ \ [ < ) ] [ / < [ > \ ) ] > [ ) : ) ( [ / \ ) ( ] / [ > ( [ : ) [ < ] : [ \ / [ ( < \ [ / > [ / [ > < [ / ) ] < > ) < ) ( [ ] [ \ ) [ / > [ ( \ > ] ) / > ) \ ( [ \ > < [ / : < [ ) > \ [ / ) > [ ( < ) / > < ) [ > [ < [ ] / [ \ [\(>[])</[><:>/<[]\)<[(>[/)\[<>[])][\:[] ) : [ > ] [ ) / [ > < [ / ) ] / < [ > \ [ / \ [ ] < / [ \ > ( [ : ) [ : ) \ ] < > : [ \ ) < / [ ) \ [ / [ > ( ) > \ < ) / [ ( ) \ > [ ] [ ) \ [ > / > [ ] \ [ < [ > [ / ) [ < > [ \ ( > [ ] ) < / [ > < ) > / < [ ] \ ) < [ ( > [ / ) \ [ < > ] ] ) ( < [ \ / ] [ < [ > ) < > [ / < [ ] : [ < > \ [ > [ \ ) < ] [ > \ / [ : < ) /<)\[]>[/)[/][:[\<)])[><[\)/[>([<)\][</> 279 H E D H B C A H G E H D F E B H I G H H G E H A C H G E H C A B H E F B C H D A H B E I F H E H F E G I H C H E I C H H A C H F B H E D H C D A G F C I H C F H B A F E A C F C H B D H F G H H C A H E F A H D H F E H B F H A D E H A E H E G D H G H B C H G H I E H C I E F H H C D B C H F D E B H E B H A H C B E H F A E H H G C H G D E H B A E H D A C H E H A E H G H D A F H B I F E H D C B E A H G C I A E H F D C H B G H A D H H B F E A G H F E H A C H D A E F H C B E B H C A F G H G B A H A F H F G E H A C H C F A H A H E U B H C B C H H E A D I H G F B B F E D C H A C H E I B H C I G H E H E H A H H D E I B C H G C F H H C H A B H D C H E H C I H C H E A H F H E A C H C B H D E C H C F I E A C H E H B F I H G C H D C D G B H C B C H B A E I H H F E H B H E C A D H A C H I G E D C H I A H / ) \ / ] [ > / ) ) / \ < ) ] / : ) / / ) ) / > [ / ) ) / [ > ] / ) < ] [ / \ > / ] ) : < / ) / < ) ) : / [ / ) : [ / / > [ / < ] / ) \ / [ \ > ) < [ : / [ < / ] > < ) > [ < [ / ] \ / < ) / / [ > / ) < > / \ / < ) / ] < / > \ ) / > ) / ) \ ) / ) / ] [ / ) / : ) / [ : ) < / / [ \ ] [ / < \ ) ] / ) ] / > / [ ] ) / < > ) / / ) [ / ) \ ) / ] > ) / \ > [ / ) / > ) / ) / \ > < / ] : < ) / \ [ ] ) > / ) [ \ > ) / < \ [ / ] ) / > \ / / ] < ) > ) / < ) / > [ / \ > ) < / [ ] ) ] / [ > < ) / ) ] > / > < / < ) ) / > [ / [ < > / > / ) \ ] / [ ] [ / / ) > \ : / ) < ] ] < ) \ [ / > [ / ) : ] / [ : ) / ) / ) / > / / \ ) : ] [ / ) [ < / / [ / > ] / \ [ / ) / [ : / [ / ) > / < / ) > [ / [ ] / \ ) [ / [ < : ) > [ / ) / ] < : / ) [ / \ [ \ ) ] / [ ] [ / ] > ) : / / < ) / ] / ) [ > \ / > [ / : ) ) \ [ / : > / Scoring Sheet for the Reading Task Participant ID# Scoring System: (NOTE: Write down all details about participant's response in the space provided) # points Response 3 participant remembers to spell the appropriate word out loud in response to the presentation of the target cue ("res") 2 participant responds to the target cue in way that is different from what was instructed (e.g., simply saying the word aloud, or indicating that something should be done); 1 participant responds to the wrong cue (e.g., "rel"), or responds late (after the cue is no longer shown). 0 participant fails to respond at all; does not respond to the cue Classification of Errors: Type of Error I Response Serious a serious error is an omission. The participant fails to respond to the cue; simply reads the word and continues reading without remembering that he/she was supposed to respond to the cue. Intermediate participant initially reads a word beginning with "res" but then becomes aware of their mistake and quickly makes the correct response. Slight participant realizes he/she is making a mistake in the course of reading the target word and immediately corrects it (reads part of the word, stops, then spells it out). NOTE: Other relevant responses = indicate if (and W H E N ) participant asked about "res" instructions (e.g., "am I supposed to do the "res" thing now?"); indicate WHICH inappropriate cues (e.g., "receiver", "rasped") to which responses were made and what type of response was made (e.g., spelled out full word, started to spell out the word and stopped, etc.); were any comments made about specific responses?, etc., etc. ltem# Score Description/Comments about response 1. resentfully type of error: 2. resemblance type of error: 3 . resting type of error: 4. resist type of error: 5. reserve type of error: 6. restaurant type of error: 7. resonating type of error: 8. respiratory type of error: 9. responded type of error: other relevant responses: 282 Reading Task Instructions: (Hand the subject a copy of the story "Agnes" FACE DOWN, instruct participant to turn the page in their booklet, and then read the instructions aloud asking participants to follow along in their response booklet. Be sure they do not see the comprehension questions below!!) This next task is designed to test verbal fluency and text comprehension. Therefore, I'm going to ask you to read aloud the short story in front of you. Try to read it with average speed, neither very quickly nor very slowly. While reading you may, of course, make some mistakes but don't get too anxious or worried about them, just continue on with your reading. In short, just read this story as you would perhaps read a fairy tale to a child at home. I would also like you to keep in mind that your comprehension of the story will be tested at the end of the task, so you will also have to pay attention to the content of the story. This is one of the tasks in which I will audiotape you, so I can score it more accurately at a later time. ^ Start audiotape now. Instructions: Please begin reading now. Follow along with the Test key/Experimenter version of the story "Agnes." Score participants responses on the Reading Task Scoring Sheet provided. Write out all details of the participant's responses. When participant has finished reading the story, administer the following questions: (keep audiotaping) Comprehension Questions: Continue audiotaping. Administer these questions using the participant's response booklet. Ask the following three questions aloud. Let the participant answer the questions verbally and write down his/her answers in the space provided in the participant's response booklet. 1. Who was Agnes in this story? 2. What were the names of the people who the elderly lady was gossiping about? 3. At the end of the story, what did the elderly lady do when she called the wrong number again? 283 FollOW-Up Questions: (Again, administer these questions using the participant's response booklet. Read all questions along with participant and point to the rating scales where applicable. Be sure to cover over questions falling below the one being asked at the time. Write down participant's answers for him/her). 1. Were you told to do something extra while you were reading the story? (a) if yes, what were you supposed to do? (b) if no, were you supposed to look for certain letters or words while you were reading? 2. Take a look at the following three-letter combinations. Which three letters were you asked to look for at the beginning of words while reading the story? (circle correct answer) rep rem ret rea res rel rec 3. Was it difficult to perform simultaneously two tasks like reading the story and spelling out all words that started with the letters "res"? 1 2 3 4 Not Difficult Slightly Difficult Quite Difficult Very Difficult 4. Did you constantly think about the task to spell out words while reading or did you remember it only when you encountered words beginning with "res"? Circle one: Y e s , I constantly thought about it No, I only thought about it when I encountered "res" Comments/Deta i ls : 5a. How interesting for you was the story you were reading? 1 2 3 4 Not Interesting Slightly Interesting Quite Interesting Very Interesting 5b. Did you have some text-unrelated thoughts during the reading, i.e., to what extent were you absorbed in reading? 1 2 3 4 Not Absorbed Slightly Absorbed Quite Absorbed Very Absorbed 284 6. How willing or motivated were you to spell out words beginning with "res" while reading the story? 1 2 3 4 Not Willing Slightly Willing Quite Willing Very Willing 7. Rate how well you think you performed in responding to the target "res" according to the following scale: 1 2 3 4 Never Some of the time Often Every Time 8. How confident are you about your answer in the previous question, question 7? 1 2 3 4 Not Confident Somewhat Confident Quite Confident Very Confident 9. Give a numerical estimate of the total number responses you made in the reading task. In other words, how many times did you spell out a word beginning with the letters "res" while reading the story? 10. Estimate the number of times you responded to "res" in total, for all the tasks so far. 285 "Agnes"--Scoring Scheme for the Comprehension Questions for the Reading Task Score: (maximum score = 6 pts.) Question 1. Who was Agnes in this story? Pts. Criteria (sample answers) 2 pts. --The person the old battle-axe was trying to call - T h e unknown person who the old lady was attempting to reach by telephone --The person the main character was pretending to be at the other end of the phone line 0 pts. - T h e old battle-axe --The elderly lady who kept calling the wrong number - T h e person at the other end of the phone line - T h e main character (all other incorrect answers) Question 2. What were the names of the people who the elderly lady was gossiping about? Pts. Criteria (sample answers) 2 pts. - A n y TWO of the following names: Walter, Ella, George 1 pt. - A n y ONE of the above names 0 pts. - N o n e of the above names (e.g., Gert, Madge, Eddie, etc.) Question 3. At the end of the story what did the elderly lady do when she called the wrong number again? Pts. Criteria (sample answers) 2 pts. 1 pt. for E A C H of the following: - S a i d she had the wrong number (1 pt.) -S lammed down or hung up the phone (1 pt.) 1 pt. ONE of the above responses 0 pts. None of the above responses 286 North American Adult Reading Test (Set the list of words in front of the participant. The list is on the next page. Follow along using scoring/pronunciation key provided. Audiotape responses and score response as you go along). Instructions: The next test measures your ability to pronounce a variety of different words from the English language. I want you to read slowly down this list of words starting here (point to debt) and continuing down this column and on to the next. When you have finished reading the words on the page, turn the page over and begin here (indicate top of second page). After each word please wait until I say "Next" before reading the next word. I must warn you that there are many words that you probably won't recognize, in fact, most people don't know them, so just guess at these, O.K.? This is the other task in which I will audiotape your performance. (Encourage participants to guess. Participants are allowed to change a response if they wish, however, if more than one is given they must make a decision. All responses should be reinforced with "good", "that's fine", etc.). ^Start audiotape when participant is ready. Instructions: Go ahead. f^Stop audiotape when participant has finished. North American Adult Reading Test (NAART) (Blair & Spreen, 1989) DEBT DEBRIS AISLE REIGN DEPOT SIMILE LINGERIE RECIPE GOUGE HEIR SUBTLE CATACOMB BOUQUET GAUGE COLONEL SUBPOENA PLACEBO PROCREATE PSALM BANAL RAREFY GIST CORPS HORS D'OEUVRE SIEVE HIATUS GAUCHE ZELOT PARADIGM FAQADE CELLIST INDICT DETENTE IMPUGN CAPON RADIX AEON EPITOME EQUIVOCAL REIFY INDICES ASSIGNATE TOPIARY CAVEAT SUPERFLOUS LEVIATHAN PRELATE QUADRUPED SIDEREAL ABSTEMIOUS BEATIFY GAOLED DEMESNE SYNCOPE ENNUI DRACHM CIDEVANT EPERGNE VIVACE TALIPES SYNECDOCHE 289 Memory Questionnaire (from R. Hannon, University of the Pacific, Stockton, California) Instructions: The following questionnaire has been developed to test how well you remember to do things. Please answer each question to the best of your knowledge. For each item, select the place on the line which best indicates your behaviour during the past week or month or year, as indicated below each item. The questions vary according to whether week, month, or year is specified, so pay close attention. Circle the slash (| ) you select, as demonstrated below: Example: I forgot to water my plants: never 2 times/ 4 or more month times/month The person responding to the above question forgot to water his/her plants approximately 3 times during the past month. If the item does not apply to you during the time specified, circle NA just to the right of the item (for not applicable). For example, if you do not own any plants, you would respond as demonstrated below: I forgot to water my plants: JA" never 2 times/ 4 or more month times/month Again, be sure to respond to each item. Thank you very much for your cooperation! 290 1. I missed appointments I had scheduled: never 3 times/ month •— | NA 6 or more times/month 2. I forgot to follow a change in my usual routine: I I never 2 times/ month .... | NA 4 or more times/month 3. I forgot to send a card for a birthday or anniversary: never 3 times/ month | | NA 6 or more times/month 4. I forgot to make an important phone call: never 2 times/ week .... | NA 4 or more times/week 5. I told someone something that I did not mean to tell: never 2 times/ month .... | NA 4 or more times/month 6. I forgot to return something I borrowed: never 2 times/ month •— | NA 4 or more times/month 7. I forgot to pick up items I needed when shopping: never 2 times/ week •— | NA 4 or more times/week 291 8. I forgot to meet a friend on time: never 2 times/ week .... | NA 4 or more times/week 9. I forgot to pass on a message to someone: never 2 times/ week I NA 4 or more times/week 10. I forgot to run an errand I mean to do: never 3 times/ week .... | NA 6 or more times/week 11. I forgot to return a phone call: I I I I I never 2 times/ week — | NA 4 or more times/week 12. I forgot to make an appointment I needed to make (e.g., doctor or dentist): never 2 times/ month •— | NA 4 or more times/month 13. I forgot to write an important letter: never 2 times/ month .... | NA 4 or more times/month 14. I forgot to return books to the library by the due date: never 2 times/ month •— | NA 4 or more times/month 292 15. I forgot to tip when I finished dinner at a restaurant: never 2 times/ month •— | NA 4 or more times/month 16. I forgot to turn my alarm clock off when I got up in the morning: never 2 times/ week —-1 NA 4 or more times/week 17. I forgot to lock the door when leaving my apartment or house: never 2 times/ week — | NA 4 or more times/week 18. I forgot to take my keys out of my car before locking the doors: never 2 times/ month .... | NA 4 or more times/month 19. I forgot to button or zip some part of my clothing as I was dressing: never 2 times/ week •— | NA 4 or more times/week 20. I forgot to pay the bill when finishing a meal at a restaurant: I — never 2 times/ month — | NA 4 or more times/month 21. I forgot to put a stamp on a letter before mailing it: never 2 times/ month — | NA 4 or more times/month 293 22. I forgot to comb my hair in the morning: never 2 times/ 4 or more week times/week 23. I forgot to put on deodorant after showering or bathing: never 2 times/ 4 or more week times/week 24. I forgot to flush the toilet: never 2 times/ week | | •— | NA 4 or more times/week 25. I forgot to get the groceries out of the car when I got home from the grocery store: never 2 times/ 4 or more month times/month 26. I forgot to lock up my house, bike, or car: never 2 times/ 4 or more week times/week 27. I forgot to shower or bathe: never ' ' ' 2 times/ 4 or more week times/week 28. I forgot to cash or deposit my paycheck before my account ran out of money: never ' ' 2 times/ 4 or more month times/month 294 29. I forgot what I wanted to say in the middle of a sentence: never 2 times/ week •— | NA 4 or more times/week 30. I forgot to say something important I had in mind at the beginning of a conversation: I I never 2 times/ week | 1 „ — | NA 4 or more times/week 31. I forgot what I came into a room to get: - I - I never 2 times/ week | | NA 4 or more times/week 32. I started to do something, and then forgot what it was I wanted to do: never 2 times/ week — | NA 4 or more times/week 33. I forgot to bring something I meant to take with me when leaving the house never 2 times/ month | | NA 4 or more times/month 34. I got part way through a chore and forgot to finish it: never 2 times/ week — | NA 4 or more times/week 35. I was driving and temporarily forgot where I was going: never 2 times/ month — | NA 4 or more times/month 295 36. I dialed someone on the phone and forgot who I called by the time they answered: never . . . . . | 2 times/ month I | . . . . . •— | NA 4 or more times/month 37. I started writing a note or letter and forgot what I wanted to say: | | never 2 times/ month | . . . . . — | NA 4 or more times/month 38. I started to write a cheque and forgot to whom it was to be paid: never 2 times/ month — | NA 4 or more times/month 39. I make lists of things I need to do: never 2 times/ week — | NA 4 or more times/week 40. I write myself reminder notes: | | | | . . . never 2 times/ week — | NA 4 or more times/week 41. I make a grocery list whenever I go shopping for food: never 2 times/ week — | NA 4 or more times/week 42. I plan my daily schedule in advance so I will not forget things: never 2 times/ week .... | NA 4 or more times/week 296 43. I repeat things I need to do several times to myself in order to remember: never 2 times/ 4 or more week times/week 44. I use external reminders like tying a string around my finger to help me remember to do things: never 2 times/ 4 or more week times/week 45. I rehearse things in my mind so I will not forget to do them: never 2 times/ 4 or more week times/week 46. I lay things I need to take with me by the door so I will not forget them: never 2 times/ 4 or more week times/week 47. I make Post-It (sticky notes) reminders and place them in obvious places: | | | | | | | | | NA never 2 times/ 4 or more week times/week 48. I create mental pictures to help me remember to do something: never 2 times/ 4 or more week times/week 297 49. I put things in piles so I know which ones to do first and which can wait: never 2 times/ 4 or more week times/week 50. I lay in bed at night and think of things I need to do the next day so I won't forget to do them: never 2 times/ 4 or more week times/week 51. I try to do things at a regular time so I will remember to do them: never 2 times/ 4 or more week times/week 52. I keep an appointment book updated in order to remember to do things: never 2 times/ 4 or more week times/week 298 DES Eve Bernstein Carlson, Ph.D. & Frank W. Putnam, M.D. (1993) Directions: This questionnaire consists of twenty-eight questions about experiences that you may have in your daily life. We are interested in how often you have these experiences. It is important, however, that your answers show how often these experiences happen to you when you are not under the influence of alcohol or drugs. To answer the questions, please determine to what degree the experience described in the question applies to you and circle the number to show what percentage of the time you have the experience. Example: 0% 10 20 30 40 50 60 70 80 90 100% (never) (always) Date: Age: Gender: M F 1. Some people have the experience of driving or riding in a car or bus or subway and suddenly realizing that they don't remember what has happened during all or part of the trip. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 2. Some people find that sometimes they are listening to someone talk and they suddenly realize that they did not hear part or all of what was said. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 3. Some people have the experience of finding themselves in a place and having no idea how they got there. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 4. Some people have the experience of finding themselves dressed in clothes that they don't remember putting on. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 5. Some people have the experience of finding new things among their belongings that they do not remember buying. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 6. Some people sometimes find that they are approached by people who they do not know who call them by another name or insist that they have met them before. Circle a number to show what percentage of time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 7. Some people sometimes have the experience of feeling as though they are standing next to themselves or watching themselves do something and they actually see themselves as if they were looking at another person. Circle a number to show what percentage of time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 299 8. Some people are told that they sometimes do not recognize friends or family members. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 9. Some people find that they have no memory for some important events in their lives (for example, a wedding or graduation). Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 10. Some people have the experience of being accused of lying when they do not think they have lied. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 11. Some people have the experience of looking in a mirror and not recognizing themselves. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 12. Some people have the experience of feeling that other people, objects, and the world around them are not real. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 13. Some people have the experience of feeling that their body does not seem to belong to them. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 14. Some people have the experience of sometimes remembering a past event so vividly that they feel as if they were reliving that event. Circle the number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 15. Some people have the experience of not being sure whether things that they remember happening really did happen or whether they just dreamed them. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 16. Some people have the experience of being in a familiar place but finding it strange and unfamiliar. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 17. Some people find that when they are watching television or a movie they become so absorbed in the story that they are unaware of other events happening around them. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 18. Some people find that they become so involved in a fantasy or daydream that it feels as though it were really happening to them. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 19. Some people find that they sometimes are able to ignore pain. Circle a number to show what percentage of the time this happens to you. 0% 10 20 30 40 50 60 70 80 90 100% 300 20. Some people find that they sometimes sit staring off into space, thinking of nothing, are not aware of t