@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Arts, Faculty of"@en, "Psychology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Enright, Corinne S."@en ; dcterms:issued "2010-09-04T00:31:18Z"@en, "1989"@en ; vivo:relatedDegree "Master of Arts - MA"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The following thesis reviews literature relevant to vivid autobiographical memories. In particular, attention is paid the theoretical debate over whether a special memory mechanism is required to explain the existence of vivid memories (VMs). The three experiments presented in this thesis addressed four main questions. First, are some life-events are more likely to be vividly recalled than others? Second, can the "Now Print" special memory mechanism theory proposed by Brown and Kulik (1977) explain the existence of VMs? Third, can ordinary memory theories explain the existence of VMs? Fourth what aspects of distinctiveness allow the memory system to differentiate between recall of events in a vivid versus non-vivid manner? The first question was explored by categorizing reported memories into thematic clusters and comparing the percentage of VMs falling into each category with a control group of readily available memories in Experiment 1, and to a control group of non-VMs in Experiment 2. Both experiments found that the percentage of VMs reported for injurious and life-threatening events was significantly higher in the VM groups than in either control group. The second question was addressed by examining the relationships between vividness and the two primary factors (consequentiality and affective arousal) that Brown and Kulik proposed as triggers for the special mechanism. Although both of these factors were significantly correlated with vividness, the nature of the relationship between these factors and vividness was not found to be of the discontinuous and necessary nature required by the "Now Print" model. Further, the relationship between consequentially and vividness was found in Experiment 2 to be explainable in terms of rehearsal. In order to address the third question, subjects were asked, in both Experiments 2 and 3, to rate their memories on scales which indexed factors generally thought to underlie superior memory performance (rehearsal and arousal). Each of these factors were found to be significantly correlated with vividness, however, they could not explain a great deal of the variance in vividness ratings. The fourth issue addressed in this thesis, is whether distinctiveness of certain aspects of the event serve to differentiate between well-remembered non-VMs and well-remembered VMs. Distinctiveness of affective and cognitive aspects of the event were found to be significantly associated with vividness of recall, whereas, other measures of distinctiveness were not. Also discussed in this thesis are the relationship between event pleasantness and vivid recall, sex differences and content effects."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/28217?expand=metadata"@en ; skos:note "CHARACTERISTICS UNDERLYING VIVID AUTOBIOGRAPHICAL MEMORIES By CORINNE S. ENRIGHT B.A., The University of B r i t i s h Columbia, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1989 © Corinne S. Enright, 1989 '5' / 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 j s c ^ hoU^ y The University of British Columbia Vancouver, Canada Date J u t . X>6+Vff DE-6 (2/88) V i v i d Autobiographical Memories i i Abstract The following thesis reviews l i t e r a t u r e relevant to v i v i d autobiographical memories. In p a r t i c u l a r , attention i s paid the t h e o r e t i c a l debate over whether a s p e c i a l memory mechanism i s required to explain the existence of v i v i d memories (VMs). The three experiments presented i n t h i s t h e s i s addressed four main questions. F i r s t , are some l i f e - e v e n t s are more l i k e l y to be v i v i d l y r e c a l l e d than others? Second, can the \"Now P r i n t \" s p e c i a l memory mechanism theory proposed by Brown and Kulik (1977) explain the existence of VMs? Third, can ordinary memory theories explain the existence of VMs? Fourth what aspects of d i s t i n c t i v e n e s s allow the memory system to d i f f e r e n t i a t e between r e c a l l of events i n a v i v i d versus non-v i v i d manner? The f i r s t question was explored by categorizing reported memories into thematic c l u s t e r s and comparing the percentage of VMs f a l l i n g into each category with a control group of r e a d i l y a v a i l a b l e memories i n Experiment 1, and to a control group of non-VMs i n Experiment 2. Both experiments found that the percentage of VMs reported for injurious and l i f e -threatening events was s i g n i f i c a n t l y higher i n the VM groups than i n e i t h e r control group. The second question was addressed by examining the r e l a t i o n s h i p s between vividness and the two primary factors (consequentiality and a f f e c t i v e arousal) that Brown and Kulik V i v i d Autobiographical Memories i i i proposed as t r i g g e r s f o r the sp e c i a l mechanism. Although both of these factors were s i g n i f i c a n t l y correlated with vividness, the nature of the r e l a t i o n s h i p between these factors and vividness was not found to be of the discontinuous and necessary nature required by the \"Now P r i n t \" model. Further, the r e l a t i o n s h i p between c o n s e q u e n t i a l l y and vividness was found i n Experiment 2 to be explainable i n terms of rehearsal. In order to address the t h i r d question, subjects were asked, i n both Experiments 2 and 3, to rate t h e i r memories on scales which indexed factors generally thought to underlie superior memory performance (rehearsal and arousal). Each of these factors were found to be s i g n i f i c a n t l y correlated with vividness, however, they could not explain a great deal of the variance i n vividness ratings. The fourth issue addressed i n t h i s t h e s i s , i s whether d i s t i n c t i v e n e s s of c e r t a i n aspects of the event serve to d i f f e r e n t i a t e between well-remembered non-VMs and w e l l -remembered VMs. Distinctiveness of a f f e c t i v e and cognitive aspects of the event were found to be s i g n i f i c a n t l y associated with vividness of r e c a l l , whereas, other measures of di s t i n c t i v e n e s s were not. Also discussed i n t h i s thesis are the r e l a t i o n s h i p between event pleasantness and v i v i d r e c a l l , sex differences and content e f f e c t s . , V i v i d Autobiographical Memories i v Table of Contents Abstract i i Table of Contents i v L i s t of Tables ' v i L i s t of Figures v i i Acknowledgements v i i i Introduction 1 Attri b u t e s of V i v i d Memories 1 Statement of Aims 4 E x i s t i n g Theories •••• 4 Theoretical Differences 8 Discontinuity Versus Continuity 8 Persistence and Accuracy 9 Underlying Factors 14 Thematic Content 2 3 Methodological Issues 2 3 Operationally Defining the Phenomenon ... 25 Variable Manipulation 3 0 Cueing 3 0 Control Groups 3 2 V e r i f i c a t i o n 33 L i t e r a t u r e Review 34 Accuracy 3 4 Persistence 41 Consequentiality 42 Novelty 4 6 V i v i d Autobiographical Memories v Surprise 47 A f f e c t 48 Distinctiveness 49 Rehearsal 50 Arousal 52 Thematic Content 52 Li t e r a t u r e Review Conclusions 53 Overview of Study 55 Experiment 1 56 Method 59 Results 61 Discussion 67 Experiment 2 71 Method 75 Results 7 6 Discussion 83 Experiment 3 85 Method 89 Results 91 Discussion 99 General Discussion 102 References 109 Appendix A 122 Appendix B 125 Appendix C 128 Appendix D 132 V i v i d Autobiographical Memories v i L i s t of Tables Table 1. Experiment 1. T-tests between pleasant and unpleasant event memories 112 Table 2. Experiment 2. T-tests between v i v i d and non-v i v i d memories 113 Table 3. Experiment 2. Pearson correlations 114 Table 4. Experiment 2. P a r t i a l c o rrelations c o e f f i c i e n t s with vividness 115 Table 5. Experiment 3. T-test analyses 116 Table 6. Experiment 3. Correlational Analyses 117 V i v i d Autobiographical Memories v i i L i s t of Figures Figure 1. Experiment 1. Thematic category percentages 118 Figure 2. Experiment 2. Thematic category percentages 119 Figure 3a. Experiment 2. Scatterplot of rehearsal with v i v i d 120 Figure 3b. Experiment 3. Scatterplot of rehearsal with v i v i d 121 V i v i d Autobiographical Memories v i i i Acknowledgements I would l i k e to thank my supervisor Peter Graf and my committee members John Y u i l l e and Brian De Vries for t h e i r assistance and patience. I would also l i k e to thank Stan Coren, Lawrence Ward, Jim Russell, Del Paulus, Sherri Hancock and Trudy Berschied for t h e i r invaluable assistance and support. F i n a l l y , I would l i k e to thank my family, p a r t i c u l a r l y my s i s t e r Cathy for showing me t h i s could be done. V i v i d Autobiographical Memories 1 Factors Underlying V i v i d Autobiographical Memories People experience countless events over the course of t h e i r l i v e s and the perceptual q u a l i t i e s of the memories for these events vary greatly. For some events, even those recently experienced, i t i s d i f f i c u l t to produce anything more than hazy, p a r t i a l r e c o l l e c t i o n s of what might have occurred at a c e r t a i n time. Other events, however, seem to be r e c a l l e d c l e a r l y , completely, and immediately even a f t e r several years have past. These l a t t e r memories, which have been termed \"flashbulb memories\" by some investigators and \" v i v i d memories\" by others, have only recently begun to be systematically studied. At t r i b u t e s of V i v i d Memories The f i r s t formal inves t i g a t i o n of v i v i d memories was conducted i n 1899 by F. W. Colgrove. Colgrove interviewed 179 middle-aged and older subjects, asking them \"Do you r e c a l l where you were when you heard that Lincoln was shot?\" Even though t h i r t y - t h r e e years had passed since the assassination of President Lincoln, 71% of the subjects were able to provide, at minimum, information about where they were when they f i r s t heard the news, what time of day i t was, and who t o l d them the news. Many were able to provide much more detaile d , v i v i d accounts of the event. Colgrove concluded that memories for \" v i v i d experiences\" have an abiding c h a r a c t e r i s t i c . V i v i d Autobiographical Memories 2 Brown and Kulik (1977) conducted t h e i r own updated version of Colgrove's study. They asked subjects to r e c o l l e c t personally experienced events of a shocking nature, as well as the circumstances i n which they f i r s t heard about various c u l t u r a l l y important events, such as the assassinations of John F. Kennedy, Robert Kennedy, Martin Luther King, and Malcolm X. Brown and Kulik were impressed by t h e i r subjects' a b i l i t i e s to give detailed, perceptually v i v i d accounts of many events and they c a l l e d them \"flashbulb memories\". Like Colgrove, Brown and Kulik noted the unusual persistence and d e t a i l of flashbulb memories; however, unlike Colgrove, who was concerned only with persistence, Brown and Kulik also emphasized the content and perceptual q u a l i t y of the memories. They explained: John Kennedy was shot t h i r t e e n years ago . . . Almost everyone t e s t i f i e s that h i s r e c a l l of h i s circumstances i s not an inference from a regular routine. I t has a primary ' l i v e ' q u a l i t y that i s almost perceptual. Indeed, i t i s very l i k e a photograph that indiscriminately preserves the scene i n which each of us found himself when the flashbulb was f i r e d , (p. 74) They defined flashbulb memories as \"memories for the circumstances i n which one f i r s t learned of a very s u r p r i s i n g and consequential (or emotionally arousing) event\" (p. 73). Rubin and Kozin (1984) questioned Brown and Kulik's contention that only very surprising, consequential events produce highly v i v i d r e c o l l e c t i o n s . They suggested instead that flashbulb memories were simply a s p e c i a l l y defined subset V i v i d Autobiographical Memories 3 of v i v i d l y r e c a l l e d events that d i f f e r from the larger set only by v i r t u e of t h e i r being s u r p r i s i n g and consequential. In order to study t h i s question they asked subjects to record t h e i r three c l e a r e s t , most v i v i d , l i f e l i k e , autobiographical memories. Unlike Brown and Kulik's flashbulb memories, these r e c o l l e c t i o n s were not required to be associated with s u r p r i s i n g emotional or consequential events. Rubin and Kozin also asked t h e i r subjects to rate t h e i r r e c o l l e c t i o n s and the events associated with them on several dimensions including national and personal importance, consequentiality, and surprise. Based on analyses of these ratings, Rubin and Kozin argued: . . . that unless there e x i s t s some properties of flashbulb memories that are d i f f e r e n t from v i v i d memories, the two categories should not be kept separate. The only d i f f e r e n t properties, according to Brown and Kulik's d e f i n i t i o n and the data c o l l e c t e d here, are consequentiality and surprise. These two properties, however, may not be s u f f i c i e n t to warrant a separate category; they were included by d e f i n i t i o n and have yet to receive support outside the d e f i n i t i o n i t s e l f , (p. 92) Rubin and Kozin concluded that flashbulb memories are indeed a s p e c i a l l y defined subset of v i v i d memories. Further, they pointed out that the dis t i n g u i s h i n g features of a flashbulb memory comes from the event that gives r i s e to i t and there i s nothing s p e c i a l about i t s memory representation. For the purposes of t h i s paper, the term flashbulb memories w i l l be used i n the sense intended by Brown and Kulik. That i s , flashbulb memories (FBs) are perceptually V i v i d Autobiographical Memories 4 v i v i d , p e r s i s t e n t memories that are associated with novel, consequential, emotional events. The term v i v i d memory (VM) w i l l be used to r e f e r to the larger set of v i v i d l y r e c a l l e d , p e r s i s t e n t memories regardless of the c h a r a c t e r i s t i c of the event they r e f e r to. The t h e o r e t i c a l implications associated with each of these terms w i l l be discussed i n further d e t a i l . Statement of Aims The purpose of the studies presented i n t h i s t h e s i s was to investigate and describe the c h a r a c t e r i s t i c s of VMs and the events associated with them. In discussing these studies, s p e c i f i c attention w i l l be paid to the following questions: (a) Are some l i f e events more l i k e l y to be v i v i d l y r e c a l l e d than others?; (b) What i s the r e l a t i o n s h i p between memory vividness and consequentiality, novelty, surprise and a f f e c t ? ; (c) Are other factors such as arousal, d i s t i n c t i v e n e s s , and rehearsal also r e l a t e d to vividness of r e c a l l ? ; and (d) Is the pattern of re l a t i o n s h i p s among these factors more consistent with a s p e c i a l memory mechanism view or with ordinary memory theory? Before d i r e c t l y addressing these questions, e x i s t i n g theories and related research w i l l be discussed. E x i s t i n g Theories The v i v i d q u a l i t y and unusual persistence of c e r t a i n memories i s not generally disputed. The mechanisms that underlie the formation and maintenance of such memories, V i v i d Autobiographical Memories 5 however, does not enjoy the same immunity from controversy. Two main l i n e s of thought are evident i n the l i t e r a t u r e . The f i r s t was proposed by Brown and Kulik (1977) who suggested that FBs are the product of a s p e c i a l memory mechanism that i s triggered by novelty, consequentiality, and emotional arousal. Brown and Kulik's flashbulb mechanism was modeled a f t e r Robert B. Livingston's (1967) \"Now P r i n t ! \" mechanism. Livingston postulated a five-step neurological process that begins with \"Reticular recognition of novelty\" and \"Limbic discrimination of b i o l o g i c a l meaning f o r the i n d i v i d u a l \" and r e s u l t s i n \" a l l recent brain events, a l l recent conduction a c t i v i t i e s \" being \"printed\" (p. 576). Brown and Kulik's non-neurological i n t e r p r e t a t i o n of t h i s model i s that upon experiencing an event: . . . f i r s t comes the recognition of high novelty or unexpectedness; then comes a t e s t for b i o l o g i c a l meaning f o r the i n d i v i d u a l [consequentiality]; i f t h i s second t e s t i s met, there follows the permanent r e g i s t r a t i o n not only of the s i g n i f i c a n t novelty, but of a l l recent brain events, (p. 77) A strong version of such a memory theory has several questionable implications for the nature of FBs, which Neisser (1982) summarizes as follows: This hypotheses implies (a) that flashbulb memories are accurate; (b) that the process by which the memory i s created occurs at the time of the event i t s e l f ; (c) that surprise, emotionality, and s i m i l a r reactions are c l o s e l y correlated with the \"consequentiality\" of an event, and that higher l e v e l s of surprise and emotionality lead to good memory; and (d) that the s i m i l a r i t i e s among d i f f e r e n t flashbulb memories r e f l e c t the common c h a r a c t e r i s t i c s of an underlying neural mechanism. (P- 43) V i v i d Autobiographical Memories 6 A l i t e r a l i n t e r p r e t a t i o n of the FB analogy, therefore, would require that r e c a l l be e n t i r e l y accurate, complete, permanently encoded at the time of the event, triggered by aspects of the event (consequentiality and su r p r i s e ) , and formed by a s p e c i a l i z e d FB memory mechanism. I t should be pointed out, however, that Brown and Kulik did not make such a strong claim f o r t h e i r s p e c i a l mechanism. At l e a s t , concerning the question of completeness of encoding, Brown and Kulik q u a l i f i e d t h e i r flashbulb analogy, claiming that: Flashbulb memory . . . i s a good name f o r the phenomenon i n as much as i t suggests surprise, an indiscriminate illumination, and brevity. But the name i s inappropriate i n one respect that had better be brought forward at once. An actual photograph, taken by flashbulb, preserves everything within i t s scope; i t i s altogether indiscriminate. Our flashbulb memories are not. . . . In short, a flashbulb memory i s only somewhat indiscriminate and i s very f a r from complete, (pp. 74-75) Brown and Kulik d i d not, however, make any suggestions as to what c r i t e r i a or mechanisms might govern t h i s discrimination between those d e t a i l s that are permanently encoded and those that are not. I t i s perhaps because they were not s p e c i f i c about such important aspects of FBs, that Brown and Kulik's view has often been portrayed and c r i t i q u e d i n the l i t e r a t u r e (such as i n Neisser, 1982) as though they had proposed a strong version of the sp e c i a l mechanism theory. There are, i n fact, no true proponents of the strong version of t h i s s p e c i a l mechanism, yet, i t i s the strong version (inferred from a l i t e r a l i n t e r p r e t a t i o n of the flashbulb analogy) that i s most V i v i d Autobiographical Memories 7 often discussed i n the l i t e r a t u r e . Although t h i s could be viewed as s e t t i n g up a straw person for the purpose of argument, t h i s strategy has been r a t i o n a l i z e d by McCloskey, Wible, and Cohen (1988) as follows: Regardless of what Brown and Kulik intended, i t i s possible to consider strong claims about c h a r a c t e r i s t i c s of flashbulb memories because . . . these claims would provide the firmest basis f o r asserting that flashbulb memories are very d i f f e r e n t from ordinary memories, (p. 172) Whether a strong (indiscriminate) or weak (discriminate) version of the FB hypothesis i s being considered w i l l a f f e c t the relevance of questions about memory, accuracy, and persistence that are most often seen as central to the special/ordinary memory mechanism controversy. This t o p i c w i l l be discussed i n greater d e t a i l i n subsequent sections of t h i s paper. The necessity for a spe c i a l mechanism (weak or strong) i n explaining FBs has been questioned by several authors including McCloskey, et a l . (1988), Neisser (1982) and Rubin and Kozin (1984). McCloskey et a l . (1988) suggested that: Flashbulb memories may be viewed as memories f o r s i g n i f i c a n t and d i s t i n c t i v e personal experiences, and hence as memory explicable i n terms of ordinary memory mechanisms . . . the d i s t i n c t i o n between the flashbulb memories and other sorts of autobiographic memories i s an a r t i f i c i a l and a r b i t r a r y one. (p. 181.) I f FBs and VMs are explainable i n terms of ordinary memory theories, then there would be no need to propose a separate memory mechanism. The question remains, however, whether FBs are s i g n i f i c a n t l y d i f f e r e n t from other V i v i d Autobiographical Memories 8 autobiographical memories and, i f they are, i s the difference of such a nature that a s p e c i a l mechanism would be necessary to explain i t . The next section addresses the sp e c i a l mechanism/ordinary memory controversy. Theoretical Differences Discontinuity Versus Continuity Brown and Kulik's argument that a s p e c i a l FB memory mechanism e x i s t s assumes that memories produced by t h i s mechanism are q u a l i t a t i v e l y d i s t i n c t from other memories. C h a r a c t e r i s t i c s of FB memories, therefore, would not necessarily be predicted from, or explainable within the realm of ordinary memory theories. According to Brown and Kulik, the f i r i n g of the spe c i a l mechanism i s all-or-none; i t occurs i n s i t u a t i o n s that are s u f f i c i e n t l y extreme to t r i g g e r the mechanism or else no FB w i l l be registered. The to-be-remembered event e i t h e r does or does not possess the unique q u a l i t i e s that t r i g g e r a FB memory. Unlike Brown and Kulik*s view, most memory theories conceptualize vividness as a continuous dimension of memory. FB memories may be located toward the high extreme of the v i v i d dimension but they are no more than a qu a n t i t a t i v e l y defined subset of ordinary autobiographical memories. McCloskey et a l . (1988) have summarized t h i s view as follows: V i v i d Autobiographical Memories 9 Memories fo r experiences of learning about su r p r i s i n g , consequential events are continuous i n t h e i r c h a r a c t e r i s t i c s with other autobiographical memories and conform to the same p r i n c i p l e s . . . . there i s no q u a l i t a t i v e d i s t i n c t i o n to be drawn between memories for learning about shocking, important events, and memories fo r learning about expected, t r i v i a l events, (p. 181) Formation of FB memories are, according to most memory theories, not an all-or-none process. These memories may simply possess q u a n t i t a t i v e l y more perceptual c h a r a c t e r i s t i c s r e l a t i v e to other age-comparable memories. The d i s t i n c t i o n between FBs and non-FBs i s a matter of degree. FBs need not be conceptualized as discontinuous with the ordinary memory system; they can simply be thought of as a s p e c i a l l y defined subset of v i v i d l y encoded memories. The question of whether a s p e c i a l mechanism i s required to explain FBs has been addressed by research focussing on the persistence and accuracy of memories. The strong version of the s p e c i a l mechanism view predicts that FBs are permanent and 100% accurate. Although i t i s true that permanence and e i d e t i c accuracy would be d i f f i c u l t to account for without evoking some sort of s p e c i a l theory, the absence of such permanence and accuracy does not necessarily disprove Brown and Kulik's weaker spe c i a l mechanism theory. Persistence and Accuracy In everyday language, the phrase \"I remember i t l i k e i t happened yesterday\", i s often used to describe FB memories. In t h i s sense, then FBs might be said to defy the ordinary V i v i d Autobiographical Memories 10 wear-and-tear that time i n f l i c t s on our a b i l i t y to r e c a l l events. Ordinary memories, whether based on laboratory studies or autobiographical events, consistently show a negatively accelerating function c a l l e d the Ebbinghaus fo r g e t t i n g curve that describes the general r e l a t i o n s h i p between time and the amount of information remembered. Forgetting i s not generally viewed as an actual loss of information that had previously been available, but rather, as an i n a b i l i t y to r e t r i e v e the memory or c e r t a i n aspects of i t . With enough e f f o r t and proper r e t r i e v a l s t rategies, the information might s t i l l be r e c a l l e d . FBs are remarkable because they do not appear to f i t e i t h e r the Ebbinghaus for g e t t i n g function or the usual fading of perceptual c h a r a c t e r i s t i c s . Further, they are remarkable because they seem to be r e t r i e v e d e f f o r t l e s s l y . Ordinary memory proponents explain these observations by assuming that events that produce FB memories have sp e c i a l c h a r a c t e r i s t i c s that make the memories r e l a t i v e l y more r e s i s t a n t to forgetting. \"The experience of learning about an event may be more s i g n i f i c a n t , d i s t i n c t i v e , and so f o r t h - and therefore more memorable\" (McCloskey et a l . , p. 181). In other words, FBs have f l a t t e r f o r g e t t i n g curves which may be predictable from the c h a r a c t e r i s t i c s of the o r i g i n a l event. Accuracy of r e c o l l e c t i o n has generally been viewed as an a d d i t i o n a l c h a r a c t e r i s t i c of FB memories that i s c r i t i c a l for a r b i t r a t i n g questions about the s p e c i a l mechanism hypothesis. V i v i d Autobiographical Memories 11 Consistent with the strong s p e c i a l mechanism view, i s the notion that, i f inaccuracies e x i s t , the event can not be stored i n a FB form. L i t t l e , i f any, attention has been paid to Brown and K u l i k s 1 suggestion that the s p e c i a l mechanism may discriminate between aspects of the event that are and are not encoded. A further p o s s i b i l i t y that has been overlooked, i s the idea that the sp e c i a l mechanism need not work i n t o t a l i s o l a t i o n from ordinary memory stores. Most current theories accept that memory i s reconstructive. That i s , memories are not stored as mental copies of events that occur i n the world; instead, they are thought to be stored i n cognitive frameworks (generally refer r e d to as schemata), from which the o r i g i n a l events can be reconstructed with varying degrees of accuracy. The obvious advantage of such a schema-based system, i s that i t allows f o r greater organization and integration of information while maximizing the capacity of the system to r e t a i n vast numbers of memories. Within these schematic stores, the encoding of new events can both influence and be influenced by previous event memories. In t h i s way, proactive and re t r o a c t i v e interferences, that lead to inaccuracies, can occur. Further, the act of r e t r i e v i n g the memory might also change the manner i n which the event i s subsequently r e c a l l e d . Neisser (1982) has proposed that the concept of self-schemata (schemata that pertain to our own identity) can be used to V i v i d Autobiographical Memories 12 explain the complete and apparently accurate nature of FBs, saying that: . . . the \"narrative structure\" (the who, what, when, where and why) of these reports can be explained simply as a r e f l e c t i o n of narrative convention f o r which everyone i n our culture i s at l e a s t somewhat f a m i l i a r . . . . We have a schemata for the a r r i v a l of important news . . . and we apply these schemata when we t r y to remember how we heard about President Kennedy's assassination. . . . we remember the d e t a i l s of a flashbulb occasion because those d e t a i l s are the l i n k s between our own h i s t o r i e s and \"History\". . . . We rehearse the occasion often i n our minds and our conversations, seeking some meaning i n i t . Indeed, i n a good novel there would be some meaning to such a sudden eruption of the public into the private; something l a t e r i n the story would hinge on i t . In r e a l l i f e there may be no such meaning, but we cannot help seeking i t . The more we seek i t , the more compelling our memory of the moment becomes. That memory may be accurate—frequent rehearsal and discussion probably contribute to accuracy—but i t need not be. I t s purpose i s served equally well whether or not the d e t a i l s are correct. I t i s the very existence of the memory that matters, not i t s contents, (pp. 47-48) The apparent accuracy of FBs, according to Neisser, has no necessary r e l a t i o n s h i p to our actual r e c a l l of the event, but i s simply a product of memory processing. In other words, FBs are a natural consequence of how our reconstructive memory works. Such a s t r i c t r e c onstructionist view i s not held by a l l autobiographical memory t h e o r i s t s . Brewer (1986, 1988) has proposed a p a r t i a l reconstructive theory that i s more op t i m i s t i c about our a b i l i t y to accurately r e t a i n information about our l i v e s . Brewer's view i s that: V i v i d Autobiographical Memories 13 . . . recent personal memories r e t a i n a r e l a t i v e l y large amount of s p e c i f i c information from the o r i g i n a l phenomenal experience (eg. location, point of view) but that with time, or under strong schema-based processes, the o r i g i n a l experience can be reconstructed to produce a new non-veridical personal memory that retains most of the phenomenal c h a r a c t e r i s t i c s of other personal memories. (1986, p.44) According to Brewer's view, inaccuracies i n autobiographical memories are a function of time. A completely accurate account of what you had f o r breakfast t h i s morning might not be remarkable; however, the same degree of accuracy f o r a breakfast a month, a year, or several years ago would bear some consideration. A highly detailed, accurate account of the instance of f i r s t hearing about a p r e s i d e n t i a l assassination that occurred t h i r t e e n years ago would c e r t a i n l y merit the same consideration. I t was because of the apparent accuracy of FBs that Brown and Kulik rejected the idea that a reconstructive theory could explain FBs. The s p e c i a l mechanism can be best described as a copy theory. Copy theories hold that memories are v e r i d i c a l representations of the o r i g i n a l event. The strong version of t h i s theory holds that FB memories, l i k e photographs, capture and preserve every d e t a i l of the o r i g i n a l event. The weaker version, however, holds that the mechanism discriminates against encoding some aspects of the o r i g i n a l event. These missing pieces might be f i l l e d i n by schema-based processes and thus inaccuracies, transformations, and confusions could r e s u l t . These f i l l e d - i n items may gradually become non-V i v i d Autobiographical Memories 14 re t r i e v a b l e from the ordinary memory stores, thus producing a s l i g h t l y negative accelerated forgetting curve. Such a forg e t t i n g function would be very d i f f i c u l t to d i s t i n g u i s h from the Ebbinghaus curve predicted by ordinary memory theories. Whereas evidence of inaccuracies i n FB memories may serve to cast doubt on the strong s p e c i a l mechanism theory, i t does not necessarily damage the argument fo r Brown and Kulik's discriminating mechanism. Although examination of the rel a t i o n s h i p s between accuracy and persistence, and factors such as novelty, consequentiality and vividness of r e c a l l are i n t e r e s t i n g i n t h e i r own r i g h t , they can not answer the c r i t i c a l question of whether FBs are formed by a s p e c i a l , discontinuous, mechanism or are explainable within ordinary autobiographical memory theories. Much more could be learned about the mechanisms underlying the formation, storage, and r e t r i e v a l of FB memories by examining the r e l a t i o n s h i p between FBs and factors generally thought to boost remembering. This issue w i l l be discussed i n the following sections. Underlying Factors Various suggestions have been made about the factors that underlie the existence of VMs. Although there i s some overlap between s p e c i a l mechanism and ordinary memory theories regarding what these factors are, there are also several important differences. The p r i n c i p l e differences have to do V i v i d Autobiographical Memories 15 with event c h a r a c t e r i s t i c s that give r i s e to FB memories and the memory mechanism i t s e l f . The s p e c i a l mechanism theory requires that i f , and only i f , c r i t i c a l l e v e l s of consequentiality, novelty, surprise, and a f f e c t are experienced, the Now Pr i n t mechanism w i l l be triggered and a FB w i l l r e s u l t . Therefore, a l l events giv i n g r i s e to FB memories would, by necessity, have to be consequential, novel, surprising, and emotional. In contrast, ordinary memory theories generally hold that the q u a l i t y of a memory representation can be affected at a c q u i s i t i o n , during storage, and at r e t r i e v a l . D i f f erent factors could influence processing at any of these stages, thereby contributing to vividness of subsequent remembering. The factors suggested by Brown and Kulik (consequentiality, novelty, surprise, and affect) may be i n d i r e c t l y related to memory vividness but only by v i r t u e of t h e i r e f f e c t on the qu a l i t y of a c q u i s i t i o n , the i n t e g r i t y of storage, and the ease of r e t r i e v a b i l i t y . Three factors known to have a strong influence on ac q u i s i t i o n , storage, and r e t r i e v a l , respectively; are arousal, d i s t i n c t i v e n e s s , and rehearsal. According to McCloskey et a l . , I t i s commonly accepted that the strength and d u r a b i l i t y of memories produced by ordinary memory mechanisms are influenced by factors such as the di s t i n c t i v e n e s s of the to-be-remembered material, the i n t e r e s t [arousal] and s i g n i f i c a n c e i t holds for the i n d i v i d u a l , and the extent to which i t i s rehearsed. I f , then, experiences i n which one learns of a surprising, consequential event are indeed more s i g n i f i c a n t , d i s t i n c t i v e , or i n t e r e s t i n g V i v i d Autobiographical Memories 16 than the commonplace experiences of everyday l i f e , or more often rehearsed, there i s reason to expect that ordinary mechanisms would y i e l d considerably better memories for the former than the l a t t e r , (p. 180) Although arousal, d i s t i n c t i v e n e s s and rehearsal are i n t e r r e l a t e d to some extent, each i s thought to influence a d i f f e r e n t stage of memory and thus, each w i l l be discussed separately. Arousal. What did Brown and Kulik mean by arousal? They used the term frequently i n connection with consequentiality, as i n the following l i n e s from t h e i r abstract: The p r i n c i p a l two determinants [for the occurrence of a FB] appear to be a high l e v e l of surprise, a high l e v e l of consequentiality, or perhaps emotional arousal . . . I f these two variables do not a t t a i n s u f f i c i e n t l y high l e v e l s , no flashbulb memory occurs, (p. 73) Within ordinary memory theories, arousal i s used to r e f e r to a general l e v e l of alertness and a c t i v a t i o n of the body. An inverted, U-shaped function has been found to describe the re l a t i o n s h i p between arousal and performance on a wide range of tasks. This r e l a t i o n s h i p , c a l l e d the Yerkes-Dodson law (1908), states that there i s an optimal degree of arousal f o r any task. One p o s s i b i l i t y consistent with t h i s law, i s that VMs may r e s u l t when events are experienced under optimal l e v e l s of arousal. I f t h i s i s the case, then memories associated with extremely non-arousing or with extremely arousing events may not be v i v i d l y r e c a l l e d . Events that f a l l within an optimal range may be better attended to and, V i v i d Autobiographical Memories 17 therefore, more f u l l y processed at the time of encoding. Arousing events also tend to be d i s t i n c t i v e . The a b i l i t y to di s t i n g u i s h an event i n memory from s i m i l a r events might also a f f e c t the q u a l i t y of i t s memory representation. Distinctiveness . An event that i s d i s t i n c t i v e , r e l a t i v e to other events i n memory, possesses aspects which do not e a s i l y f i t into e x i s t i n g cognitive structures. These unique aspects, once brought to mind, can serve as cues to r e c a l l other aspects of the event. Therefore, a d i s t i n c t i v e event i s more l i k e l y to be r e c a l l a b l e as an i n d i v i d u a l instance. D i s t i n g u i s h a b i l i t y of events should give an advantage to memories fo r events that mark the f i r s t time, the l a s t time, the best time, or the only time something occurred. In fact, anything about the event that would render i t le s s schematizable, more i n d i v i d u a l , should make i t more memorable. The l i t e r a t u r e has l a b e l l e d t h i s the \"von Restorff e f f e c t \" ; i t describes the fac t that events that are d i s t i n c t i v e are well remembered (von Restorff, 1933). Rehearsal. Within autobiographical memory, rehearsal i s usually defined as how often we think about or discuss an event. I t i s viewed as f a c i l i t a t i n g encoding, maintaining memories, and rendering them easier to r e t r i e v e . The causal sequence i s generally seen as going from rehearsal to memory. In other words, higher l e v e l s of rehearsal cause memories to become or remain more v i v i d l y and elaborately encoded. V i v i d Autobiographical Memories 18 Brown and Kulik proposed that t h i s r e l a t i o n s h i p between rehearsal and memory does not necessarily hold f o r FBs. Because the encoding of these memories i n e i t h e r the strong or weak version of the s p e c i a l mechanism theory i s thought to be all-or-none, rehearsal could neither f a c i l i t a t e nor augment the encoding process. Further, because the memory was already permanent, rehearsal would not be necessary to maintain the memory representation. I f there i s a r e l a t i o n s h i p between rehearsal and FBs, Brown and Kulik saw the causal d i r e c t i o n as going from memory to rehearsal. They explained as follows: Probably, the rehearsal process, set o f f by high consequentiality, draws i t s content from the unchanging FB memory, but i n rehearsal, a verbal narrative i s l i k e l y to be created. . . . We propose that rehearsals b u i l d up associative strength between the verbal narrative created and the r e t r i e v a l cues used i n the various settings, (p. 86) Any r e l a t i o n s h i p , therefore, between FBs and high l e v e l s of rehearsal i s seen, i n t h i s view, as simply r e f l e c t i n g the f a c t that both might be related to a t h i r d factor such as consequentiality. I t makes i n t u i t i v e sense to assume that we think about and discuss consequential events more than non-consequential events. Repeated discussions or thoughts about the event, Brown and Kulik propose, may subsequently produce more elaborated and s t y l i z e d reporting of the event but would have no d i r e c t e f f e c t on the vividness of the memory. In other words, repeated t e l l i n g of an event improves the t e l l i n g , but not the memory representation. V i v i d Autobiographical Memories 19 I f rehearsal appears to be a necessary requirement f o r FB memories then t h i s would c e r t a i n l y cast doubt on the need to propose a s p e c i a l mechanism. On the other hand, i f no r e l a t i o n s h i p e x i s t s between rehearsal and FB memories, t h i s would not necessarily support the idea that a s p e c i a l mechanism e x i s t s because the memories* perceptual q u a l i t i e s could be the r e s u l t of t h e i r o r i g i n a l a c q u i s i t i o n and storage. In other words, arousal, d i s t i n c t i v e n e s s , and rehearsal may a l l be s u f f i c i e n t but not necessary factors f o r d i s t i n c t i v e encoding. In the next few sections Brown and Kulik's proposed underlying factors of consequentiality, novelty, surprise, and a f f e c t w i l l be discussed by focussing on t h e i r influence on remembering, and t h e i r r e l a t i o n s h i p to ordinary memory theories. Consequentiality. Brown and Kulik used the word \"consequentiality\" to describe events that have b i o l o g i c a l s i g n i f i c a n c e f o r the rememberer. Upon reviewing Brown and Kulik's studies, Neisser (1982) c r i t i c i z e d the idea that consequentiality could t r i g g e r a FB response because the long term e f f e c t s of an event are often unknown u n t i l some time a f t e r the event has taken place. At the time of an event, however, i t i s reasonable to presume that people have some perception about how consequential an event may be. This perceived consequentiality, therefore, could serve as the t r i g g e r i n g mechanism. Making such a judgment may require V i v i d Autobiographical Memories 20 nothing more than u t i l i z i n g Neisser's \"schema for important news\". In general, memory theories would predict a p o s i t i v e c o r r e l a t i o n between consequentiality and v i v i d r e c a l l . This r e l a t i o n s h i p could be the r e s u l t of perceived consequentiality (at the time of the event) or actual consequentiality (at the time of r e c a l l ) . High expected consequentiality could a f f e c t the q u a l i t y of the o r i g i n a l encoding. Further, events that remain consequential tend to be thought about and discussed (rehearsed) more often. Novelty and surprise. The sp e c i a l mechanism theory s t i p u l a t e s that events giving r i s e to FBs are novel. In other words, the event must pass some minimum c r i t i c a l requirement for being s u f f i c i e n t l y unlike other events that the i n d i v i d u a l had experienced up u n t i l that time. Ordinary memory theories also suggest that novel events are better r e c a l l e d than events that the rememberer has experienced several times. This \"novelty advantage\" may a f f e c t memory at the a c q u i s i t i o n , storage, and r e t r i e v a l stages. Novel events may be more arousing than non-novel events. Further, f i r s t - t i m e events are, by d e f i n i t i o n , d i s t i n c t i v e because there can only be one f i r s t f or any cl a s s of events. Related to novelty, Brown and Kulik suggested that surprise may also be an element necessary to the f i r i n g of the FB mechanism. Therefore, a l l events giving r i s e to FBs must be s u r p r i s i n g . Ordinary memory theories do not see surprise as a necessary condition for v i v i d encoding. Conway and V i v i d Autobiographical Memories 21 Berkerian (1988) suggested that, not only i s surprise unnecessary, but, i n some cases, pre-event a n t i c i p a t i o n may serve to prime memory so that v i v i d encoding of an expected event might be f a c i l i t a t e d . For example, weddings and graduations are not generally s u r p r i s i n g events, yet such events can c e r t a i n l y be v i v i d l y r e c a l l e d . Any p o s i t i v e r e l a t i o n s h i p between surprise and vividness of encoding might be explained within ordinary memory theories, because s u r p r i s i n g events are generally d i s t i n c t i v e and arousing. A f f e c t . Brown and Kulik have suggested that a f f e c t i s also r e l a t e d to the formation of FBs. What i s unclear, however, i s whether the important t r i g g e r i n g element i s a f f e c t i n t e n s i t y (degree of pleasantness or unpleasantness) or the general arousal usually associated with intensely a f f e c t i v e s i t u a t i o n s . The influence that arousal might have on memory qu a l i t y has already been discussed. Ordinary memory theories suggest that the a f f e c t i v e tone of an event may help to d i s t i n g u i s h i t from other s i m i l a r events. Events that are extremely unpleasant or extremely pleasant would be distinguishable from events that are within the ordinary range of emotional experiences. To the degree that the a f f e c t associated with an event i s d i s t i n c t i v e , the event might be more v i v i d l y r e c a l l e d . Although Brown and Kulik focused t h e i r study on unpleasant events, there i s nothing i n the theory that presupposes that the FB mechanism i s more l i k e l y to be V i v i d Autobiographical Memories 22 triggered by unpleasant events than by pleasant events. S i m i l a r l y , within the ordinary memory l i t e r a t u r e , although i t has been suggested that differences i n r e t r i e v a b i l i t y may a f f e c t the r e l a t i v e reporting of pleasant versus unpleasant memories ( i e , Psychoanalytic theory and mood s p e c i f i c memory theor i e s ) , there i s no t h e o r e t i c a l reason to suggest differences i n vividness of r e c a l l based on pleasantness or unpleasantness of the event experienced. The special/ordinary memory controversy cannot be meaningfully discussed by looking at any p a r t i c u l a r aspect of an event but must be considered i n l i g h t of the o v e r a l l r e l a t i o n s h i p between event c h a r a c t e r i s t i c s and remembering performance. Both views agree that high l e v e l s of consequentiality, arousal, and novelty would lead to more v i v i d l y encoded, e a s i l y r e t r i e v a b l e memory representations. The s p e c i a l mechanism theory predicts that these factors are a l l necessary to the formation of a v i v i d representation. Ordinary memory theories, on the other hand, view these factors as being related to vividness of r e c a l l because of t h e i r r e l a t i o n s h i p to arousal, d i s t i n c t i v e n e s s , and rehearsal. Some clues may be given as to the mechanisms underlying the formation of FBs by looking, not simply at the i n d i v i d u a l factors that are thought to give r i s e to v i v i d memories, but at the thematic content of memories that are v i v i d l y r e c a l l e d . V i v i d Autobiographical Memories 23 Thematic Content I f c e r t a i n l i f e events consistently do and others do not produce FB memories, then an examination of the c h a r a c t e r i s t i c s of these events may illuminate the processes that underlie the formation of FBs. According to the s p e c i a l mechanism view (weak or strong), some c r i t i c a l l e v e l of both novelty and consequentiality must be present i n order to t r i g g e r the FB mechanism. Therefore, FB's should neither e x i s t f o r expected events, such as graduations and weddings, nor f o r events whose occurrence was experienced as mundane or inconsequential even i f they were l a t e r (even moments later) r e a l i z e d to be novel or of consequence. Ordinary memory theories p r e d i c t that a wide range of events w i l l produce v i v i d memories based on a v a r i e t y of factors such as arousal, d i s t i n c t i v e n e s s , and rehearsal. Higher l e v e l s and a greater number of these factors would produce r e l a t i v e l y higher l e v e l s of vividness. The thematic content of subjects' VMs could provide a great deal of information about the factors that underlie v i v i d r e c a l l and may be useful i n determining whether a s p e c i a l mechanism i s necessary to explain t h i s phenomenon. Before reviewing the l i t e r a t u r e relevant to t h i s area, methodological issues w i l l be discussed. Methodological Issues Colgrove's o r i g i n a l investigation into FB memories was based on \"A well known pedagogical p r i n c i p l e . . . that v i v i d V i v i d Autobiographical Memories 24 impressions are e a s i l y r e c a l l e d \" (p. 247). His purpose f o r running the study was to t e s t \"the abiding character of a v i v i d experience\" (p. 247). Colgrove chose the event of f i r s t hearing about the assassination of President Lincoln as the v i v i d experience f o r h i s study. No explanation of why t h i s event was chosen was provided. Personal interviews were used to obtain memory accounts and subjects* responses were c l a s s i f i e d as e i t h e r affirmative or negative. An affi r m a t i v e account required that the subject be able to provide facts \"as to the time of day, exact location, and who t o l d them.\" The en t i r e reporting of t h i s study covers a mere four paragraphs and concludes as follows: Of the 179 persons interviewed, 127 r e p l i e d i n the affir m a t i v e , and were able to give f u l l p a r t i c u l a r s ; 52 r e p l i e d i n the negative. A few who gave a negative reply r e c a l l e d where they were when they heard of Gar f i e l d ' s death. Inasmuch as 33 years have elapsed since Lincoln's death the number who made an affi r m a t i v e reply must be considered large, and bears testimony to the abiding character of v i v i d experiences, (p. 247) By the research standards of present day experimental psychology, Colgrove's study leaves a great deal to be desired. Where are the operational d e f i n i t i o n s , the control groups, the s t a t i s t i c s ? S t i l l , t h i s study has the advantage of being able to say something about a phenomenon that could not be studied under a r t i f i c i a l , contrived laboratory conditions. Memories f o r v i v i d r e a l - l i f e experiences can only be meaningfully studied within a relevant r e a l - l i f e context. V i v i d Autobiographical Memories 25 The advent of the s c i e n t i f i c study of memory i s generally i d e n t i f i e d with the 1885 publication of Ebbinghaus's monograph \"Uber das Gedachnis\". As t h i s new science became more and more entrenched i n the laboratory, i t i s not s u r p r i s i n g that studies such as Colgrove*s disappeared from the l i t e r a t u r e . A renewed i n t e r e s t i n r e a l - l i f e memory phenomena and concern with e c o l o g i c a l v a l i d i t y has been evident over the past two decades (for discussion see Neisser, 1982, 1988; Winograd, 1988). There has been, as Winograd (1988) points out, \"a tendency to view the ecol o g i c a l approach to memory as contrasting sharply i n both method and theory with the Ebbinghaus t r a d i t i o n \" (p. 11). This, Winograd claims, i s a mistake. The challenge for researchers interested i n any form of r e a l - l i f e memory, i s to r e t a i n as many of the advantages of laboratory studies without compromising the ec o l o g i c a l v a l i d i t y of the phenomenon being studied. Therefore, methodological concerns such as operationally defined v a r i a b l e s , v a r i a b l e manipulations, cueing procedures, control groups, and accuracy v e r i f i c a t i o n are of equal concern to non-laboratory studies as to laboratory studies. Each of these issues w i l l be discussed i n turn. Operationally Defining the Phenomenon One of the f i r s t problems, i n any type of research, i s to operationally define the dependant vari a b l e . When scanning our own memories, i t i s f a i r l y easy for us to imagine the type V i v i d Autobiographical Memories 26 of memory referred to as FBs. Establishing a c r i t e r i o n (or set of c r i t e r i a ) f o r determining whether someone else's memory q u a l i f i e s as a FB i s , however, not as simple and there does not, as yet, appear to be a d e f i n i t i v e method fo r separating FBs from non-FBs. A v a r i e t y of c r i t e r i a have been used by d i f f e r e n t researchers. I t i s important when comparing r e s u l t s from d i f f e r e n t studies to consider the manner i n which FBs or VMs are defined. Differences i n r e s u l t s may simply r e f l e c t the s t r i c t n e s s or l a x i t y of the d e f i n i t i o n a l c r i t e r i a . D e f i n i t i o n by event. Colgrove's choice of an assassination of a prominent p o l i t i c a l figure as a v i v i d experience set a precedent that has been followed by several researchers (Brown & Kulik, 1977; Linton, 1975; Winograd & K i l l i n g e r , 1983). Unlike Colgrove, j u s t i f i c a t i o n f o r the choice of such an event i s generally provided by the researcher. For example, Brown and Kulik chose to t e s t memory fo r the ten events used i n t h e i r study for the following reasons: We had two major i n t u i t i o n s about the determinant of FB's when we designed our study. . . . Perhaps the most obvious property of President Kennedy's assassination was i t s extreme unexpectedness; i n most of our l i v e s no other major p o l i t i c a l figure had been assassinated. And, i n se l e c t i n g events to use i n prospecting for FB's, we generalized t h i s property and so chose ten very unexpected or novel events, among which assassinations loom large. V i v i d Autobiographical Memories 27 The second i n t u i t i o n we l u c k i l y had was that among national events l i k e assassinations, there might well be a difference between white Americans and black Americans, i n the public figures who set o f f FB's. . . .We guessed that what would matter would be the comparative consequentiality f o r the black and white i n d i v i d u a l of each national event, (pp. 76-77) A c r i t i c a l issue that arises with t h i s manner of defining FBs, i s that i t begs the question. The researcher asks subjects f o r memories of events that the researcher has already judged to possess q u a l i t i e s X and Y. The researcher then defines the responses given as FBs, and concludes that q u a l i t i e s X and Y are determinants of FBs. Further, t h i s method of defining FBs i s questionable because the researchers may have m i s - i d e n t i f i e d the c r i t i c a l components of the events, perhaps a t t r i b u t i n g importance to i r r e l e v a n t c h a r a c t e r i s t i c s , while f a i l i n g to note the importance of others. Another problem involved with defining FBs i n terms of the event, i s that i t says nothing about the q u a l i t y of the memory, only about the events associated with the memory. I f what we are studying i s a q u a l i t a t i v e memory phenomenon, then a meaningful d e f i n i t i o n of that phenomenon should r e f e r to the q u a l i t y of the memory. Memory quality , however, i s not s o l e l y determined by an event, but i s the r e s u l t of an i n t e r a c t i o n between the rememberer and the event. Therefore, an event that i s perceived by the researcher to be a v i v i d experience may not be perceived as such by the subject. For these reasons, V i v i d Autobiographical Memories 28 further c r i t e r i a are usually used to ensure that the subject's memory i s t r u l y a v i v i d one. D e f i n i t i o n by r e c a l l . Colgrove's c r i t e r i o n f o r a v i v i d memory was that the subject be able to r e c a l l at minimum information about time, l o c a t i o n and informant. Brown and Kulik established a s i m i l a r c r i t e r i o n by requiring that subjects provide information i n at l e a s t one of s i x canonical categories: place, ongoing event, informant, own a f f e c t , a f f e c t i n others, and aftermath. Other researcher (McCloskey, et a l . , 1982; Pillemer, 1984) have used modified versions of these same c r i t e r i a despite d i f f i c u l t i e s with t h i s type of d e f i n i t i o n . The major d i f f i c u l t y with t h i s type of d e f i n i t i o n i s f i n d i n g a way of meaningfully i n t e r p r e t i n g i t . I t would be hard to imagine a memory formed eith e r by a s p e c i a l or ordinary memory mechanism that contained no information from any of these categories. Setting some minimum number of categories from which information must be r e c a l l e d , however, would be a r b i t r a r y and would not u s e f u l l y d i s t i n g u i s h between FB and non-FB memories. On the other hand, providing information from a l l of these categories would not ensure that the memory had, as suggested by Brown and Kulik, the perceptual a t t r i b u t e s that characterize r e a l - l i f e events. I f FBs are thought to be phenomenally d i f f e r e n t from ordinary memories, then a more useful c r i t e r i o n for defining FBs would be the phenomenal quality, or vividness of the memory. V i v i d Autobiographical Memories 29 D e f i n i t i o n by memory qual i t y . One of the most d i r e c t methods of ascertaining whether a subject's memory q u a l i f i e s phenomenonally as a FB memory, i s to ask the subject how v i v i d the memory i s . This type of method was employed by Rubin and Kozin (1984). They f i r s t provided t h e i r subjects with a d e f i n i t i o n such as: A flashbulb memory occurs when your brain takes a pict u r e of an event. You have p a r t i c u l a r l y v i v i d memories of these events long a f t e r they occur. You tend to remember your surroundings i n exceptional d e t a i l , (p.85) Subjects were then asked to decide whether t h e i r r e c o l l e c t i o n of several events, including the assassination attempt on the l i f e of President Ronald Reagan, matched t h i s d e s c r i p t i o n . While t h i s method does assess the qu a l i t y of the memory, t h i s method produces only dichotomous r e s u l t s and, thus, can provide no ins i g h t into whether relevant factors are rela t e d to the memory q u a l i t y i n a continuous or discontinuous manner. Other researchers (eg. McCloskey, et a l . , 1988; Pillemer, 1984), therefore, have asked subjects for graduated ratings of vividness, or c l a r i t y of r e c a l l . In reviewing the l i t e r a t u r e , i t i s important to keep i n mind that the term FB memory has been used by some researchers to r e f e r to memories for c e r t a i n events, others use the term to r e f e r to the v i v i d , perceptual qu a l i t y of the memory regardless of the event being r e c a l l e d . The d e f i n i t i o n varies according to the research question asked and the manner i n which the memories are e l i c i t e d . V i v i d Autobiographical Memories 30 Variable Manipulation Laboratory studies of memory most often provide the subject with the to-be-recalled material, then have the subject give e i t h e r free or cued r e c a l l of the presented material. When dealing with r e a l - l i f e events, i t i s often not possible to present the to-be-recalled material. This i s p a r t i c u l a r l y true with FB memories because many of the vari a b l e s of i n t e r e s t (eg. a f f e c t , surprise, consequentiality) are not e a s i l y staged i n the laboratory and would be d i f f i c u l t to simulate meaningfully within e t h i c a l l i m i t a t i o n s . Events that are comparable to President Kennedy's assassination i n consequentiality, a f f e c t i v e i n t e n s i t y , and unexpectedness, could not be meaningfully nor e t h i c a l l y simulated. Therefore, FB memory studies have usually employed n a t u r a l l y occurring events that are ei t h e r defined by the experimenter or events rated by the subject as having the perceptual q u a l i t i e s or associated c h a r a c t e r i s t i c s of i n t e r e s t . In FB memory studies, the cueing procedure used can be the central means of c o n t r o l l i n g the type of memories c o l l e c t e d and therefore these procedures bear c a r e f u l consideration. Cueinq The types of questions that a memory researcher can explore depends to a great extent on the cueing method used. One of the most popular methods for accessing FBs, i s to ask people f o r t h e i r memories of na t i o n a l l y or i n t e r n a t i o n a l l y V i v i d Autobiographical Memories 31 important events. Using memories for public events has several methodological advantages. I t provides a cueing procedure that allows experimenters to e l i c i t memory accounts f o r s p e c i f i c events that occurred on a known date (and therefore are of a known retention i n t e r v a l ) , were experienced by a large population of subjects, and, i n some cases, are v e r i f i a b l e . Several studies have found FB memories for national events, such as the assassination of John F. Kennedy (Linton, 1975; Winograd & K i l l i n g e r , 1983), the f i r s t moon landing, the resignation of President Richard Nixon (Winograd & K i l l i n g e r , 1983), the assassination attempt on President Ronald Reagan (Pillemer, 1984), and the Challenger explosion (McCloskey, et a l . , 1988). FBs are, however, not confined, by d e f i n i t i o n , to events of national importance. Cued-recall studies have revealed FB memories f o r s a l i e n t personal events such as the b i r t h of a s i b l i n g (Sheingold & Tenney, 1982), f i n a l Ph.D. o r a l exams (Strube & Neubauer, 1988), and the witnessing of a crime ( Y u i l l e & Cutshall, 1986). Although using memories f o r personal events does not provide as uniform a sample as memories f o r public events, studying memories for personal events does not r e s t r i c t the range of l i f e events that can be sampled. There i s also some evidence to suggest that personal events may give r i s e to memories which are perceptually more v i v i d than public events. Larsen (1988) compared vividness ratings for memories of newsworthy reported events to ratings V i v i d Autobiographical Memories 32 for personally experienced events and found that vividness was lower f o r media-reported than personal events. Rabbitt and Winthorpe (1988) compared a procedure known as the Galton-Crovitz method, to non-cued procedures. With the Galton-Crovitz method, subjects are given a se r i e s of word cues and asked to provide a s p e c i f i c memory associated with each. Rabbitt and Winthorpe found that non-cued memory accounts were rated as more v i v i d than cued. Therefore, people's most v i v i d memories might be better accessed by simply asking subjects for unspecified v i v i d autobiographical memories. Such a procedure also has the advantage of providing information about the thematic content and temporal range of v i v i d l y r e c a l l e d events. This type of procedure would not be of much use, however, i f the purpose of the study was to access accuracy because v e r i f y i n g such a wide range of memories would, f o r a l l p r a c t i c a l purposes, be impossible. Therefore, cued procedures are generally employed f o r accuracy studies. Control Groups To date, there are only two studies i n the l i t e r a t u r e that provide a sort of non-FB control group. The f i r s t was conducted by Rubin and Kozin (1984). Subjects were cued to r e c a l l 20 events of which roughly h a l f (based on p i l o t data) would produce FB reports. Ratings of consequentiality, surprise, emotional change, and rehearsal could then be V i v i d Autobiographical Memories 33 compared between those events that were rated by the subjects to produce FB memories and those that were rated not to produce FB memories. The second control group experiment was a diary study designed by Brewer (1988). Brewer's subjects c a r r i e d a beeper-alarm with them for a period of 13 days and were instructed to write a b r i e f d escription of t h e i r actions and thoughts whenever s i g n a l l e d by the alarm to do so. At r e c a l l , subjects not only attempted to r e c o l l e c t t h e i r actions and thoughts but were also asked to rate the perceptual vividness of the memory. Therefore, the re l a t i o n s h i p between accuracy and vividness could be assessed. V e r i f i c a t i o n Because i t has generally been held that Brown and Kulik proposed a sp e c i a l mechanism that was indiscriminate, the cent r a l question i n most FB memory studies has been that of accuracy. For t h i s reason, most studies use cued procedures and further constrain the type of memory studied by requiring that some form of v e r i f i c a t i o n be possible. Two types of procedures have generally been employed to assess the accuracy of FBs. The f i r s t procedure involves finding supporting evidence, e i t h e r i n the form of documentation (Linton, 1975; Strube & Neubauer, 1988; Y u i l l e & Cutshall, 1986) or witnesses (Sheingold & Tenney, 1982; Y u i l l e & Cutshall, 1986). The second method involves a repeated measures design i n which V i v i d Autobiographical Memories 34 accounts of an event are gathered as soon as possible a f t e r the event occurs and then follow-ups are conducted at s p e c i f i e d i n t e r v a l s (Brewer, 1988; McCloskey et a l . , 1988; Pillemer, 1984; Y u i l l e & Cutshall, 1986). Accuracy of memory fo r meaningful autobiographical events i s notoriously d i f f i c u l t to t e s t . Mistakes (omissions and commissions) i n the memory protocols cannot simply be interpreted as forg e t t i n g . The subject may have been attending to some aspect of the event other than those chosen as correct by the experimenter, may not have encoded the information, or could have misinterpreted the o r i g i n a l event. A subject's r e c a l l , therefore, could be a complete and accurate accounting of the event as they perceived and encoded i t at the time of occurrence, yet not be an accurate account of the actual event. Because accuracy has received so much attention, we w i l l begin the l i t e r a t u r e review by discussing t h i s area. Literature Review Accuracy Brown and Kulik assumed that FB memories were accurate and complete accounts of whatever had been encoded at the time the event took place. Neisser's (1982) review of Brown and Kulik's s p e c i a l mechanism interpreted the theory i n i t ' s strong form and c r i t i c i z e d i t for predicting t o t a l accuracy. To support h i s contention that FB memories were not e n t i r e l y accurate, Neisser c i t e d a study by Linton. Linton (1975) had V i v i d Autobiographical Memories 35 asked her subjects simply, \"What were you doing when John Kennedy was assassinated?\" Her early r e s u l t s were reassuring. Subject's who were as young as ten years of age at the time of the assassination (ten years previously) produced memory protocols that were \" r i c h and complex\". One subject, a long-time acquaintance of Linton's, produced a protocol that although being very d e t a i l e d and expressed with great emotion and conviction, could not have been accurate. The subject claimed that i t was Linton h e r s e l f who had informed her that Kennedy had been shot. This was not only counter to Linton's own memory but was also proved impossible by documentation. Linton and her subject simply could not have been i n the same place at that time! Linton points out that: This exploration i l l u s t r a t e s again the lesson that has been learned repeatedly since the l a s t century: The researcher must know (or perhaps control) the d e t a i l s of the event being remembered. I t i s the only way to judge the accuracy of the r e c a l l , (p. 387) Neisser (1982) provided further evidence against perfect accuracy based on h i s personal FB account of hearing about the December 1941, Japanese attack on Pearl Harbor. This account included l i s t e n i n g to a baseball game on the radio. Baseball games, however, are not broadcast i n December. Further, Neisser claims \" I t can't have been a f o o t b a l l game eith e r : professional f o o t b a l l barely existed i n 1941, and then the college season ended by Thanksgiving.\" FBs, he concluded, could be \"j u s t as wrong as other kinds of memories\" and V i v i d Autobiographical Memories 36 therefore, \"are not produced by a s p e c i a l quasi-photographic mechanism\" (p. 45). Despite i t s i n t u i t i v e appeal, t h i s argument turns out to be f a u l t y . In 1986, Thompson and Cowan discovered that the news of the Pearl Harbor attack d i d interrupt a radio broadcast of a f o o t b a l l game. The game was between the New York Giants and the Chicago Bears (names we now generally associate with baseball teams) and was played at the Polo Grounds, which was the home of the New York Giants baseball team. Neisser's anecdotal evidence against the accuracy of FBs now seems more an example of s u r p r i s i n g l y accurate information about the d e t a i l s of an event that took place f o r t y years previously. The memory account produced by Linton's subject may be equally as accurate apart from the b e l i e f that Linton was the informant. Neisser (1986) responded to t h i s information with the explanation that the \"understandable\" change (substituting baseball f o r football) not only made the memory more congruent with h i s self-image but also converted i t into a personal symbol of \"Pearl Harbor Day\": . . . when my country was attacked, I was doing the q u i n t e s s e n t i a l l y American thing. [Listening to baseball] The hypothesis that the s p e c i a l status of FB's i s p r i m a r i l y due to t h e i r symbolic status predicts j u s t such transformations, (p.286) The a b i l i t y to produce instances of FB memories that contain inaccuracies does not, however, rule out the p o s s i b i l i t y that there are FBs that are e n t i r e l y accurate. V i v i d Autobiographical Memories 37 Further, the question that might be of more i n t e r e s t here, i s whether there e x i s t s a class of v i v i d l y experienced memories that are r e l a t i v e l y more r e s i s t a n t to transformations or immune to f o r g e t t i n g than non-vivid memories. Testing memory accuracy i s e s p e c i a l l y d i f f i c u l t with FBs because the experimenter i s generally not able to record the conditions i n d i v i d u a l s are exposed to i n a n a t u r a l l y occurring, s u r p r i s i n g and consequential event, and i s e t h i c a l l y r e s t r i c t e d from producing a r t i f i c i a l events of t h i s type. Two types of studies have been conducted i n an attempt to explore the question of accuracy of autobiographical memory. The f i r s t method involves finding v e r i f i c a t i o n for subjects memories of consequential l i f e events; the second involves looking at the consistency of subjects' reports over repeated t r i a l s . Sheingold and Tenney (1982) asked childr e n aged 4, 8, 12, and college students, twenty s p e c i f i c questions about a s a l i e n t childhood event, the b i r t h of a s i b l i n g . They then compared these reports with those of the subjects' mothers. They concluded that v i r t u a l l y no forgetting occurred even when the retention period was as long as sixteen years. A s i m i l a r study conducted by Strube and Neubauer (1988) asked 4 8 subjects, who had s i x months 5 years previously taken a series of f i n a l o r a l examination towards a psychology degree, to r e c o l l e c t , both i n f r e e - r e c a l l and cued r e c a l l procedures, the d e t a i l s of one s p e c i f i e d examination. They used o f f i c i a l V i v i d Autobiographical Memories 38 records to check accuracy. Thematic r e c a l l remained good (about 35%) even a f t e r three years. D e t a i l r e c a l l , however, produced a regular Ebbinghaus forgetting curve. In short, t h e i r r e s u l t s supported B a r t l e t t ' s (1932) findi n g ; people r e c a l l the g i s t of an event but not the exact d e t a i l s . V e r i f y i n g accuracy using accounts that are about the event i t s e l f but not about the subject's experiencing of the event, leaves open the question of whether or not inaccuracies are due to the subjects not having attended to the same aspects of the event that the researcher chooses to t e s t . Further, d e t a i l s that were s a l i e n t to the subjects (eg. thoughts and emotions) may be r e c a l l e d with remarkable accuracy but these aspects may not be v e r i f i a b l e . Procedures that compare subjects' i n i t i a l and follow-up accounts of experiencing the event, can eliminate t h i s question. Pillemer (1984), used t h i s type of repeated measures design to t e s t subjects' r e c a l l f o r f i r s t hearing about the assassination attempt on the l i f e of President Reagan. Using Brown and Kulik's canonical category c r i t e r i o n , Pillemer concluded that FBs for the Reagan shooting did e x i s t . Pillemer's subjects produced informationally r i c h , v i s u a l and consistent memories over the s i x month period. Factors, such as emotional reaction to the event and surprise, were found to influenced the informational complexity of the accounts not simply whether the FB mechanism f i r e s . Formation of FBs, therefore, V i v i d Autobiographical Memories 39 did not appear to be an all-or-none process, suggesting a connection with \"ordinary\" episodic memory. McCloskey et a l . (1988) conducted a s i m i l a r study comparing the v e r i d i c a l l y of subjects' memories f o r the Challenger explosion. Subjects' i n i t i a l reports were c o l l e c t e d within three days of the explosion and followed-up by a second questionnaire nine months l a t e r . Using Brown and Kulik's (1977) canonical categories c r i t e r i a , McCloskey et a l . found that a l l of t h e i r subjects had FBs of the explosion both on i n i t i a l and on follow-up t e s t i n g . Although subjects reported having high confidence that t h e i r r e c o l l e c t i o n s were accurate (average of 6.1 on a 7 point s c a l e ) , t h e i r follow-up protocols were not always consistent with t h e i r i n i t i a l accounts. Sixty-seven percent of subjects' responses were the same or more s p e c i f i c (included d e t a i l s not included i n the o r i g i n a l account) on follow-up. Nineteen percent became more general (included less d e t a i l s than the o r i g i n a l account), eight percent were inconsistent with the o r i g i n a l reports, and s i x percent of the responses were \"I don't know\". Finding s i t u a t i o n s that are surp r i s i n g and consequential and yet v e r i f i a b l e i s a major problem for researchers i n t h i s f i e l d . Y u i l l e and Cutshall (1986) were able to investigate such a s i t u a t i o n by comparing forensic evidence, i n i t i a l reports, and five-month follow-up reports for a \"shoot-out\" that was witnessed by 21 persons. They found that i n i t i a l reports were very accurate (mean score 79.3%) and that there V i v i d Autobiographical Memories 40 was l i t t l e change i n accuracy over the five-month period. Such a high degree of accuracy, they suggest: . . . i s most l i k e l y s i t u a t i o n s p e c i f i c . I t i s rare for anyone to witness a \"shoot out\" i n the middle of a busy s t r e e t i n a Canadian c i t y . The salience and uniqueness of t h i s event probably played a major r o l e i n producing v i v i d memories. Such memories may be s i m i l a r to the \"FB's\" reported by Brown and Kulik (1977), which often p e r s i s t f o r years, (p. 299) The question of whether v i v i d memories are r e l a t i v e l y more accurate than age comparable non-vivid memories, has only been studied once. Brewer (1988) used a random alarm device to s e l e c t autobiographical events over a 13 day period. Ten subjects recorded whatever they were doing and thinking immediately p r i o r to the alarm and then were tested by cued r e c a l l up to f i v e months following the event recorded. Analysis of memory ratings found that correct autobiographical r e c a l l s almost always involved high v i s u a l imagery. Although repeated measures studies provide a well c o n t r o l l e d method for accessing accuracy, for p r a c t i c a l purposes, the retention periods involved are rather short. Even i f the accuracy rates seem high, i t i s not c l e a r that there i s anything remarkable about remembering a personally experienced event with the same degree of accuracy s i x to nine months a f t e r i t occurred. I f FB memories are thought to be, not only s u r p r i s i n g l y accurate but also p e r s i s t e n t , then further follow-ups of Pillemer's and McCloskey et a l . ' s studies would c e r t a i n l y be of i n t e r e s t . V i v i d Autobiographical Memories 41 Persistence There i s some evidence that FB memories may be more persistent than non-FBs. Rubin and Kozin (1984) found that the memories t h e i r subjects rated as being FBs memories, tended to be older than those rated as non-vivid. Further, they found that the c o r r e l a t i o n between the age of the memory and the vividness r a t i n g was -.31. In other words, older memories reported i n t h i s study a c t u a l l y tended to be more v i v i d l y r e c a l l e d than newer memories. Cohen and Faulkner (1988) used the same procedure as Rubin and Kozin except that they used a larger age range (20 to 87 years) and ensured that v i v i d memories were obtained f o r every decade of the subjects' l i v e s . This was done by f i r s t asking subjects to provide s i x v i v i d memory accounts, then having them provide a memory f o r each decade of t h e i r l i v e s that was not included i n the o r i g i n a l s i x accounts. They found no s i g n i f i c a n t c o r r e l a t i o n between vividness ratings and the age of the memory, but noted that vividness declined as the age of the memory increased for about twenty to t h i r t y years and then changed l i t t l e a f t e r that. There i s some support, therefore, from memory studies of exceptional events, that accuracy under c e r t a i n s i t u a t i o n s may be s u r p r i s i n g l y high and that v i v i d memories may be r e c a l l e d over long periods of time. I f v i v i d memories are indeed more l i k e l y to be accurate and persistent, then an understanding of V i v i d Autobiographical Memories 42 the factors which give r i s e to v i v i d memory representations, i s p a r t i c u l a r l y important. Those factors proposed by Brown and Kulik (consequentiality, surprise, novelty, and affect) w i l l be reviewed f i r s t . Consequentiality Consequentiality was operationalized i n Brown and Kulik's study by having subjects rate on a fiv e - p o i n t scale how consequential the cued events were f o r them. Along with t h i s scale, subjects were provided with the following i n s t r u c t i o n s : In order to rate the consequentiality i n your l i f e of the death of someone, l e t us say President John F. Kennedy, you must t r y to imagine the things that might have gone d i f f e r e n t l y had President Kennedy l i v e d . . . . Probably the best s i n g l e question to ask yourself i n r a t i n g consequentiality i s 'what consequences f o r my l i f e , both d i r e c t and i n d i r e c t , has t h i s event had? (p. 42) Overall, the mean consequentiality r a t i n g was 2.52 f o r the ten events. However, t h i s r a t i n g gives l i t t l e information because there i s no control condition of non-FB memories with which to compare i t . Nonetheless, consequentiality ratings were found to be correlated with the number of canonical categories included i n the FB accounts (r = .79) and with the number of words i n the memory protocols (r = .88). Further, Brown and Kulik compared the number of FBs reported for events surrounding the assassinations of \" c i v i l r i g h t s leaders\" for black and white subjects. They used t h i s race-membership b c r i t e r i o n as an i n t u i t i v e measure of consequentiality because they f e l t that such events would have greater c u l t u r a l V i v i d Autobiographical Memories 43 consequences fo r blacks than for whites. The r e s u l t s showed that black subjects d i d indeed report more FBs for c i v i l r i g h t s events than did white subjects. Interestingly, Yarmey and B u l l (1987) found only s l i g h t differences between flashbulb reports by Canadian c i t i z e n s (84%) and by American c i t i z e n s (90%) even though the death of the President of the United States might not be expected to be as consequential to Canadians as i t was to American c i t i z e n s . Yarmey and B u l l did not, however, have t h e i r subjects rate consequentiality, so i t i s unclear whether Canadians did indeed perceive Kennedy's assassination as having l e s s consequences fo r them than Americans did. Other studies have included consequentiality ratings. Rubin and Kozin asked t h e i r subjects to provide three v i v i d autobiographical memories and to rate them according to how n a t i o n a l l y important, personally important, and consequential the event was to the rest of t h e i r l i v e s . The memories were rated as being of high personal importance but of low national importance. A wide range of consequentiality r a t i n g were found, therefore, high consequentiality ratings do not appear to be a necessary condition for v i v i d r e c a l l . The c o r r e l a t i o n between personal importance and consequentiality was .44 i n d i c a t i n g that, although these are perceived as r e l a t e d concepts by subjects, they are not equivalent. An event may have d i r e c t consequences that are of l i t t l e or no importance to the rememberer. However, events that are of l i t t l e d i r e c t V i v i d Autobiographical Memories 44 consequence, may s t i l l be personally important to the rememberer. Even though t h i s event had far-reaching consequences, the subjects' l i v e s continued much the same a f t e r Kennedy was assassinated as before. I t i s unclear, therefore, which aspect of t h i s concept i s being proposed as underlying v i v i d encoding. Because no control group was used i n Rubin and Kozin's f i r s t experiment, the question of whether v i v i d memories were rated as more consequential on average than non-vivid memories, could not be addressed. In order to answer t h i s and re l a t e d questions, Rubin and Kozin conducted a second study i n which subjects were cued to r e c a l l 20 events of which roughly h a l f would produce FB reports (based on r e s u l t s from a p i l o t study). FB memories were rated s i g n i f i c a n t l y higher (p < .001) on consequentiality than non-FB memories. Cohen and Faulkner (1988) had subjects rate r e t r o s p e c t i v e l y the perceived importance of the event at the time i t occurred as well as the importance the event was perceived to have a c t u a l l y had at the time of r e c a l l . Vividness was found to be more strongly correlated with ratings of importance at r e c a l l (r = .34) than with perceived importance at the time of the event (r = .22) supporting Neisser's c r i t i c i s m that consequentiality (importance) cannot t r i g g e r the FB mechanism because an event•s consequences cannot be known at the time the event takes place. V i v i d Autobiographical Memories 45 Pillemer's (1984) repeated measures approach allowed perceived consequentiality ratings to be taken shortly a f t e r the event studied (the Reagan assassination attempt) took place. A multiple regression of Pillemer's data revealed that perceived consequentiality at the time of the event made no independent contribution to memory. Therefore, the findings of Cohen and Faulkner (1988), and Pillemer (1984), cast doubt on the idea that consequentiality could serve as a t r i g g e r i n g device f o r a s p e c i a l mechanism. Further evidence that consequentiality i s not a necessary requirement f o r FBs, comes from a study conducted by Conway and Berkerian (1988). Conway and Berkerian asked subjects to provided written reports of personally experienced events; two of personal importance and two that were personally unimportant. They found that although important memories were more r e l i a b l y rated as v i v i d , both important and unimportant events produced v i v i d memories. They concluded that, because v i v i d memories f o r both important and unimportant events e x i s t , importance cannot be a necessary determinant of vividness as Rubin and Kozin suggested. Although there i s evidence that consequentiality i s related to the perceptual q u a l i t y of memories, the re l a t i o n s h i p does not appear to be of the necessary type that Brown and Kulik suggested. This r e l a t i o n s h i p between consequentiality and vividness of r e c a l l might better be explained within ordinary memory theories. V i v i d Autobiographical Memories 46 Novelty Of a l l the underlying factors proposed by Brown and Kulik, novelty has received the l e a s t attention. Cohen and Faulkner's (1988) thematic analysis suggests that novelty may be a very good predictor of an event being rated as v i v i d . Of t h e i r subjects* v i v i d personal memories seventy-three percent were unique, \"one of a kind\" events, 20% were f i r s t events ( f i r s t love, f i r s t day of school), whereas only 3% were \" l a s t time\" events. Because \"one of a kind\" events are, by d e f i n i t i o n , novel, i t would seem that Cohen and Faulkner's finding, that 93% of reported v i v i d events were novel, supports the notion that novelty may t r i g g e r a s p e c i a l mechanism. At the time an event occurs, a person cannot know whether s i m i l a r events w i l l occur. Events that were novel when they occurred, may not remain d i s t i n c t i v e . For example, Winograd and K i l l i n g e r ' s (1983) subjects were asked about t h e i r r e c a l l f o r the events surrounding f i r s t hearing of the f i r s t moon landing. Subjects' responses f o r t h i s event \"often suggested that informants were r e c a l l i n g circumstances associated with other space f l i g h t s \" (pp. 419-420). An event's being novel, therefore, does not appear to be as important as i t s being d i s t i n c t i v e . An a l t e r n a t i v e explanation for why so many of Cohen and Faulkner's subjects reported novel ( f i r s t and only) events, therefore, i s that f i r s t and l a s t events are more l i k e l y to be V i v i d Autobiographical Memories 47 d i s t i n c t i v e . Therefore, d i s t i n c t i o n rather than novelty per se, may be the relevant underlying factor. This a l t e r n a t i v e w i l l be discussed more f u l l y i n a l a t e r section. Surprise The t h i r d necessary determinant of FBs, proposed by Brown and Kulik, was surprise. This requirement was questioned by Neisser (1982) and by Conway and Berkerian (1988). I f surprise i s required, then otherwise novel and consequential events (such as weddings and birthdays) should not produce FB r e c o l l e c t i o n s . Thematic analyses of the v i v i d l y r e c a l l e d important events c o l l e c t e d from Conway and Berkerian's subjects revealed that the majority were highly predictable (weddings, b i r t h s , deaths, f i r s t jobs, leaving school, moving, changing jobs, and major l i f e changes). In Rubin and Kozin's study, subjects rated both how surprised they were when the event occurred and, from the objective point of view of an informed outsider, how l i k e l y the event was to occur. A wide range of responses was found for both of these questions i n d i c a t i n g that surprise, l i k e consequentiality and novelty, does not appear to be a necessary requirement for v i v i d r e c a l l . Further evidence that surprise i s not a necessary requirement comes from a study by Winograd and K i l l i n g e r (1983). Their subjects produced FB memories for the events surrounding t h e i r f i r s t hearing of the resignation of V i v i d Autobiographical Memories 48 President Nixon. This event was not unexpected by the American people. Although surprise does not seem to be a necessary requirement f o r FBs, i t has been found to be p o s i t i v e l y r e l a t e d to vividness of r e c a l l . Pillemer (1984) found that subjects who reported higher l e v e l s of surprise i n regard to the assassination attempt on President Reagan also produced more highly elaborated and consistent memories. Further, the VMs studied by Rubin and Kozin, were rated s i g n i f i c a n t l y higher (p_ < .001) on surprise than non-VMs. Interestingly, i n t h i s same study no s i g n i f i c a n t difference was found i n l i k e l i h o o d of occurrence ratings for FB and non-FB reports. Further, surprise and l i k e l i h o o d ratings correlated only .42 suggesting that, although these scales may be measuring a r e l a t e d concept, they are not i d e n t i c a l . Because the terms \" l i k e l i h o o d of occurrence\" and \"surprise\" share the denotative meaning of unexpectedness but \"surprise\" further suggests an element of pleasant a f f e c t i v e arousal, the r e l a t i o n s h i p between surprise and vividness of r e c a l l might be better explained by a f f e c t i v e aspects than by expectedness. A f f e c t To date, only one study has looked at the r e l a t i o n s h i p between a f f e c t and memory qual i t y . Pillemer (1984) had h i s subjects rate t h e i r \" i n t e n s i t y of emotional reaction\" experienced upon f i r s t hearing of the assassination attempt on V i v i d Autobiographical Memories 49 Reagan and to l a b e l the emotion involved. The mean rating, on a f i v e point scale, was 3.2 and the labels were usually negative (e.g., anger, sadness). Intensity of emotion did not predict memory at the one month follow-up but was s i g n i f i c a n t l y correlated with greater memory elaboration and vividness at the seven-month follow-up. Therefore, any re l a t i o n s h i p between a f f e c t i v e i n t e n s i t y and memory vividness does not appear to be immediate. Instead, a f f e c t i v e i n t e n s i t y appears to maintain the memory representation or to make i t more e a s i l y r e t r i e v a b l e . A look at the types of newsworthy events used to cue v i v i d r e c a l l (assassinations, personal shocks, national disasters) suggests that a f f e c t i v e l y unpleasant events are more l i k e l y to be v i v i d l y r e c a l l e d . This question however, has never been d i r e c t l y addressed. Having reviewed the factors underlying the existence of FBs as proposed by Brown and Kulik, i t can be concluded that, although consequentiality, novelty, surprise, and a f f e c t are rel a t e d to vividness of r e c a l l , there i s no evidence to prove that the r e l a t i o n s h i p i s a discontinuous or necessary one. The r e l a t i o n s h i p between these factors and vividness of r e c a l l may, therefore, be better explained by t h e i r r e l a t i o n s h i p to factors that are thought to a f f e c t a l l autobiographical memories; d i s t i n c t i v e n e s s , rehearsal, and arousal. V i v i d Autobiographical Memories 50 Distinctiveness As discussed e a r l i e r , Cohen and Faulkner (1988) found 93% of t h e i r subjects* v i v i d memory reports were fo r events that were novel. However, almost 80% of these novel events remained d i s t i n c t i v e \"one of a kind\" events. This casts doubt on Brown and Kulik's contention that novelty t r i g g e r s the s p e c i a l mechanism. Distinctiveness, l i k e consequentiality, cannot t r u l y be known at the time the event takes place. Therefore, i t cannot serve to t r i g g e r a FB mechanism. Cohen and Faulkner's findings suggest that the r e l a t i o n s h i p between novelty and vividness can more r e a d i l y be explained by the d i s t i n c t i v e n e s s of the event. The temporal r e l a t i o n s h i p of events to other s i m i l a r events i s not the only manner i n which events can be d i s t i n c t i v e . Events can be considered d i s t i n c t i v e because they are unlike other events within the subject's l i f e , because they are perceived to be unlike events i n other people's l i v e s , or because they involved d i s t i n c t i v e locations, a c t i v i t i e s , persons, emotions, and thoughts. None of these areas have been studied. Rehearsal Many studies (Sheingold & Tenney, 1982; Pillemer, 1984; Winograd & K i l l i n g e r , 1983) have f a i l e d to f i n d s i g n i f i c a n t c o r r e l a t i o n s between reported rehearsal and memory vividness. Rubin and Kozin asked t h e i r subjects to rate, on a fi v e - p o i n t scale, how often they had discussed the event since i t V i v i d Autobiographical Memories 51 occurred. They found a wide range of responses to t h i s question. Although FB memories were rated s i g n i f i c a n t l y higher (p. < .0001) than non-FBs memories on t h i s rehearsal r a t i n g , many v i v i d memories were reported not to be discussed at a l l . Cohen and Faulkner's (1988) r e p l i c a t i o n of Rubin and Kozin's study found low but s i g n i f i c a n t c o r r e l a t i o n s between vividness and rehearsal (accessed by three scales, frequency of r e c a l l , how often the event was discussed at the time, and how often i t i s discussed now.) Conway and Berkerian (1988), however, found a s i g n i f i c a n t c o r r e l a t i o n only with personally important memories. Evidence that rehearsal of material associated with the event might also be important for maintaining memory, comes from a study by Strube and Neubauer (1988) who found that the best predictor of high f r e e - r e c a l l scores (both thematic and de t a i l ) f o r memory of taking a f i n a l Ph.D. exam, was current professional a f f i l i a t i o n . However, t h i s factor d i d not predict recognition scores. They concluded that \"being a professional psychologist\" means that domain-specific knowledge, suited to generate e f f i c i e n t cues f o r r e c a l l , has been kept a l i v e and active. I t s cue-generating function i s not required for recognition and, accordingly, performance on the recognition task was not affected. V i v i d Autobiographical Memories 52 Arousal Although i t i s generally accepted that arousal i s related to memory performance and that FB memories are associated with arousing events, no empirical evidence e x i s t s to support t h i s . As discussed e a r l i e r , FB memories may be perceptually clearer, more persi s t e n t , and, perhaps, more accurate than non-FB memories simply because they involve l e v e l s of arousal which are optimal f o r encoding. Certainly, t h i s area i s one which merits research attention. Thematic Content Analysis of the thematic content of memories subjects report when asked for personal FBs or VMs, could give some valuable i n s i g h t s into the factors that underlie encoding, maintenance, and r e t r i e v a l of v i v i d l y - r e c a l l e d , p e r s istent memories. Rubin and Kozin derived thematic categories from memories provided by t h e i r subjects. The percentage of memories which f e l l into each of these categories suggests that some types of events may be v i v i d l y r e c a l l e d more often than others. They found that 18% of the memories were of accidents or i n j u r i e s sustained by eith e r the subject or a fr i e n d . Ten percent involved encounters with members of the opposite sex; 9% were incidents involving animals. Five percent of the memories f e l l into each of the following categories; f i r s t hearing of deaths, f i r s t week of college, and vacation experiences. Three percent were incidents at V i v i d Autobiographical Memories 53 school. T h i r t y percent of the memories were, however, non-c l a s s i f i a b l e according to Rubin and Kozin's categorization system. Cohen and Faulkner (1988) r e p l i c a t e d Rubin and Kozin's study but c o n t r o l l e d for the p r o b a b i l i t y that subjects were over-reporting events from a p a r t i c u l a r period (decade) of t h e i r l i v e s . Unlike Rubin and Kozin's thematic categorization, Cohen and Faulkner did not require that each memory f i t e x c l u s i v e l y into a single category. Twenty-three percent of the reports involved psychologically traumatic memories (deaths, quarrels, partings, being l o s t , a f r a i d , humiliated e t c . ) ; 11% involved physical trauma. Stereotypic memories f o r eight important l i f e events, accounted f o r 50% of the memories; b i r t h s , marriages, deaths (22.3%), holidays (7.3%), education (7.8%), i l l n e s s (7.7%), family (7.6%), and snapshot memories describing scenes rather than events (7.3%). Although the fa c t that FBs can be categorized into thematic groups suggests that some events are more l i k e l y to be v i v i d l y r e c a l l e d than others, the fa c t that no control group of non-FB memories was obtained i n eith e r of these studies means that the VM percentages found, could simply be representative of autobiographical memories i n general. L i t e r a t u r e Review Conclusions The s p e c i a l mechanism/ordinary memory controversy has generated several studies and commentaries. However, to date, V i v i d Autobiographical Memories 54 l i t t l e research has been directed toward t e s t i n g the r e l a t i o n s h i p between memory vividness and factors o r d i n a r i l y thought to be associated with memory qu a l i t y (rehearsal, arousal, and d i s t i n c t i v e n e s s ) . Instead, attention has been directed toward d i s c r e d i t i n g the suggestion that an indiscriminate, s p e c i a l mechanism e x i s t s . The l i t e r a t u r e does not tend to support Brown and Kulik's s p e c i a l mechanism theory. Although the factors that Brown and Kulik proposed to underlie the formation of FB memories (consequentiality, novelty, surprise, and affect) appear to be r e l a t e d to the existence of v i v i d , persistent memories, the r e l a t i o n s h i p does not appear to be of the necessary, discontinuous nature that would r e s u l t from the f i r i n g of the FB memory mechanism. The suggestion that vividness and accuracy could be p o s i t i v e l y r e l a t e d (Brewer, 1988; McCloskey et a l . , 1988; Pillemer, 1984), makes questions such as the following of greater importance. (a) Are some l i f e events more v i v i d l y r e c a l l e d than others? (b) Are v i v i d l y r e c a l l e d , persistent memories s i g n i f i c a n t l y d i f f e r e n t from age-comparable, non-vivid memories on the factors that Brown and Kulik proposed (consequentiality, novelty, surprise, and a f f e c t ) ? And i f so, i s the d i f f e r e n c e continuous or discontinuous? (c) Are v i v i d l y r e c a l l e d , persistent memories s i g n i f i c a n t l y d i f f e r e n t from age comparable, non-vivid memories on factors that are generally thought to a f f e c t memory qu a l i t y (arousal, d i s t i n c t i v e n e s s , and rehearsal)? And i f so, what i s the V i v i d Autobiographical Memories 55 nature of the relationship? (d) Is the pattern of cor r e l a t i o n s among a l l of these factors consistent with currently held autobiographical memory theories? I f the answer to t h i s l a s t question i s yes, then i t could be argued parsimoniously that a spe c i a l mechanism i s not necessary to explain the existence of VMs. I f the answer i s no, then the suggestion that there i s some form of spe c i a l mechanism (not . necessarily triggered by the factors suggested by Brown and Kulik) cannot be dismissed. Overview of Study The research presented i n t h i s thesis has been divided into three experiments. Overall, the purpose of these experiments was to address the four questions proposed i n the previous section. The f i r s t experiment explores the general c h a r a c t e r i s t i c s of v i v i d , persistent memories* thematic content, a f f e c t (pleasant vs unpleasant), and consequentiality (expected and actu a l ) . Experiment 2 addresses the question of whether or not v i v i d , persistent memories d i f f e r from non-v i v i d p e r s i s t e n t memories i n thematic content, rehearsal, consequentiality, a f f e c t , and arousal. Further, i t explores the r e l a t i o n s h i p s among vividness, rehearsal, consequentiality, a f f e c t , and arousal. The t h i r d experiment r e p l i c a t e s and expands the second by addressing add i t i o n a l factors, most notably, d i s t i n c t i v e n e s s , relevant to the spe c i a l mechanism/ordinary memory mechanism controversy. V i v i d Autobiographical Memories 56 Experiment One The main focus of the f i r s t study was to explore the thematic and a f f e c t i v e c h a r a c t e r i s t i c s of memories that people produce when asked to report VMs. The following questions were addressed. (a) What types of events give r i s e to VMs? (b) Are some types of events more l i k e l y to be v i v i d l y recalled? (c) Are VMs more l i k e l y for pleasant, unpleasant or neutral events? In addition to these questions, t h i s f i r s t study also examined the r o l e of rehearsal and consequentiality, s p e c i f i c a l l y focusing on whether consequentiality i s a necessary requirement for v i v i d and the r e l a t i o n s h i p between rehearsal and consequentiality? The experiment was run i n two phases. The purpose of the f i r s t phase was to obtain a large sample of v i v i d memory accounts that could be thematically and a f f e c t i v e l y categorized. To do t h i s , the memory protocols from a large number of subjects were c o l l e c t e d and then sorted into groups that shared s i m i l a r topics or themes. The protocols were also categorized as e i t h e r a f f e c t i v e l y pleasant, unpleasant, or neutral. While i t i s i n t e r e s t i n g to examine the frequencies with which v i v i d memories f a l l into various thematic categories, i t cannot be determined from t h i s sample whether or not the thematic d i s t r i b u t i o n of v i v i d memories i s d i s t i n c t from that of non-vivid memories. In other words, i t i s unclear i f X V i v i d Autobiographical Memories 57 percentage of the reported memories f a l l i n g into category Y t e l l s us anything sp e c i a l about VMs. I t could be that percentage X simply r e f l e c t s the percentage of events of type Y that the subjects had r e a d i l y available i n memory. Therefore, unlike previous thematic investigations of VMs (Cohen & Faulkner 1988; Rubin & Kozin, 1984), the present study also estimated the thematic d i s t r i b u t i o n of r e a d i l y a v a i l a b l e (but not necessarily vivid) memories the subjects possessed. The second phase of Experiment 1 was conducted to obtain an estimate of the number of r e a d i l y - a v a i l a b l e memories of each thematic type subjects had. Subjects were asked to estimate the number of memories that came e a s i l y to mind f o r each of the thematic categories derived from the f i r s t phase of the experiment. By comparing the percentage of reported v i v i d memories to the estimated number avail a b l e , i t can be determined whether some l i f e events are more l i k e l y to be v i v i d l y r e c a l l e d than others. I f some events are more l i k e l y than others to be v i v i d l y r e c a l l e d , then looking at c h a r a c t e r i s t i c s that are normally associated with that type of event may give some clues about factors that underlie the vividness of FB memories. A second area of i n t e r e s t i n t h i s study, was whether unpleasant or pleasant events are more l i k e l y to be v i v i d l y r e c a l l e d . I f there i s a difference i n the r e l a t i v e frequency with which pleasant and unpleasant events are represented among the reported v i v i d memories, then looking at differences V i v i d Autobiographical Memories 58 i n other factors associated with the d i r e c t i o n of a f f e c t (pleasant versus unpleasant) might also lead to a greater understanding of factors that give r i s e to VMs. A t h i r d area of i n t e r e s t i n t h i s experiment concerned the r e l a t i o n s h i p between consequentiality and v i v i d r e c a l l . Brown and Kulik's s p e c i a l mechanism view implies that events that give r i s e to VMs are, by necessity, perceived by the subjects, at the time the event takes place, to be of high expected consequence. I t i s not necessary, however, that the event a c t u a l l y turns out to be highly consequential because the VM would already be formed before the inconsequentiality of the event could be known. I f expected consequentiality i s necessary f o r the formation of VMs, then a l l VMs should be rated to have been of high expected consequentiality when they occurred, whereas, ratings of the actual consequentiality of the event could be e i t h e r high or low. On average, therefore, v i v i d memories should be rated as having higher expected consequentiality than actual consequentiality. I f consequentiality i s not necessary as a t r i g g e r f o r a s p e c i a l mechanism, then how might i t be related to v i v i d r e c a l l ? Neisser, suggested that consequential events are thought about (rehearsed) more often than non-consequential events and, therefore, the memory representation i s maintained. In order to examine t h i s p o s s i b i l i t y , the r e l a t i o n s h i p between rehearsal and consequentiality was assessed. I f rehearsal i s more c l o s e l y related to actual V i v i d Autobiographical Memories 59 consequentially than to expected consequentially, i t may be that rehearsal i s one of the factors, as Neisser suggested, that gives r i s e to VMs. On the other hand, i f there i s no difference i n the re l a t i o n s h i p between rehearsal and expected consequentiality and between rehearsal and actual consequentiality, then i t i s possible that rehearsal has no e f f e c t on VMs as Brown and Kulik proposed. Method Phase 1. Two hundred and eleven introductory psychology students (mean age 20.3 years) completed t h i s phase of the study. They had no previous formal exposure to the concept of FBs or VMs. They were tested i n a classroom s e t t i n g during regular c l a s s periods. Subjects were asked to provide three written accounts of events that had occurred more than two years ago fo r which they had \" v i v i d memories\". V i v i d memories were defined as \"memories that had perceptual q u a l i t i e s that were much l i k e r e l i v i n g the event\". Because thematic content was of primary i n t e r e s t , the experimenter was purposefully vague so as not to influence the subjects choice of memory. No examples were given. A f t e r w r i t i n g each memory account, subjects were asked to estimate how many years ago the event had taken place and to answer the following three questions using seven-point r a t i n g scales. (1) When t h i s event occurred, how much of an enduring e f f e c t did you expect i t to have on your l i f e ? (2) How much V i v i d Autobiographical Memories 60 of an e f f e c t do you f e e l t h i s event has a c t u a l l y had on your l i f e ? (3) How often do you think about t h i s event? The 633 memory protocols obtained i n t h i s phase of the study were sorted into nineteen naturally-occurring, thematic categories and labeled as being e i t h e r pleasant, unpleasant, or neutral. The categorization procedure w i l l be discussed more f u l l y i n the r e s u l t s section. Phase 2. Two weeks l a t e r , a subset of 115 Phase 1 subjects was asked to p a r t i c i p a t e i n a second phase of t h i s study. The subjects were provided with a booklet that l i s t e d the nineteen thematic categories along with several sub-categories into which an event could f a l l (see Appendix A f o r complete questionnaire). The sub-categories were meant to define each of the main thematic categories, to cue memories, and to a i d memory r e t r i e v a l by helping subjects organize t h e i r search f o r appropriate memories. The questionnaire instructed the subjects to estimate how many personal memories fo r each sub-category event came re a d i l y to mind. They were to record t h i s estimate by writing a number in d i c a t i n g approximately how many memories of each type they could think of. Before subjects' began to f i l l i n the questionnaire, the experimenter v e r b a l l y instructed them to keep the main category heading i n mind, so that i f a memory f e l l into a sub-category but did not f i t the main category they were not to count i t . Further, they were instructed that events that could be counted i n more than one category should be placed i n the category that best V i v i d Autobiographical Memories 61 f i t i n order that no memory be counted more than once. Subjects were allowed one h a l f hour to complete the questionnaire. Results The main goal of Experiment 1 was to determine the nature of events that give r i s e to VMs. For t h i s purpose, the subjects' memory protocols from Phase 1 were f i r s t sorted into 19 c l e a r l y distinguishable categories (the categories are l i s t e d i n Figure 1). The number, nature, and boundaries of these categories were determined i n t u i t i v e l y ; they were, i n the opinion of the primary investigator, the categories that permitted the best sorting of the sample c o l l e c t e d i n t h i s study. To assess the v a l i d i t y of the sorting, a collaborator was provided with the same category names (and i n some cases a b r i e f d e f i n i t i o n of the category) and asked to independently sort the memories. Inter-rater r e l i a b i l i t y was 98%. The 6 memories that were found to be problematic involved cases where the event could f a l l into more than one category. These cases were e a s i l y resolved through discussion. Both researchers also independently l a b e l l e d the protocols as expressing pleasant, unpleasant, or neutral a f f e c t . Agreement for a f f e c t i v e ratings was only 90%. The 28 cases that were problematic involved memories that contained both pleasant and unpleasant aspects, as can be seen i n the following example: V i v i d Autobiographical Memories 62 The [car] door flew open as we went around a curve and I f e l l out on my head. We went to the h o s p i t a l with me crying a l l way, not because i t hurt (actually i t was kind of fun except for the pain) but because I got a l o t of attention. I l o s t l o t s of h a i r and had scraped arms and scalp but I got a l l the attention I could ever want. Memory reports of t h i s sort were placed i n a s p e c i a l category fo r ambiguous memories. With the addition of t h i s new category i n t e r - r a t e r agreement became 100%. The ratings of avai l a b l e memories from Phase 2 of Experiment 1 were used to determine what percentage f e l l into each of the nineteen thematic categories. For t h i s purpose the percentage was f i r s t computed per thematic category for each subject, then these percentages were averaged across subjects. The f i n a l averages provided a baseline against which to compare the percentage of VMs i n each thematic category. These baseline percentages are also presented i n Figure 1 along with the percentages of VMs produced i n each category. Thematic c l u s t e r i n g of memories. The most important finding, summarized i n Figure 1, i s that the memory protocols from the f i r s t phase of Experiment 1 f a l l into c l e a r l y categorizable thematic c l u s t e r s . The figure h i g h l i g h t s systematic differences i n the frequency of VMs reported for the d i f f e r e n t categories. The most frequently reported events have to do with pleasant adult landmarks (14.1%), followed by accidents, i n j u r i e s , and i l l n e s s e s (10.6%), and f i r s t hearing about or witnessing deaths (8.9%). The smallest categories V i v i d Autobiographical Memories 63 were b i r t h s of family members (1.6%) and unpleasant adult landmarks (1.9%). Therefore, some l i f e events appear to produce many VMs whereas other l i f e events appear to produce very few. A complete account of VMs should be able to accommodate these findings. There are two possible explanations for these findings. F i r s t , the r e l a t i v e frequency with which some events are reported may indicate that there i s a s p e c i a l c h a r a c t e r i s t i c ( s ) associated with these events that gives r i s e to v i v i d r e c a l l . A l t e r n a t i v e l y , the frequency with which VMs are reported f o r each category may simply r e f l e c t the number of memories the subjects have ava i l a b l e . In order to d i s t i n g u i s h between these p o s s i b i l i t i e s , a chi-square was calcu l a t e d between the percentage of produced VMs i n Phase 1 and the percentage of rated, a v a i l a b l e (baseline) memories from Phase 2. Overall the chi-square analysis of v i v i d and baseline memories across a l l categories, revealed a s i g n i f i c a n t d i f f e r e n c e between the two memory samples (chi-square = 162.6, p_ < .001). A Pearson c o r r e l a t i o n between v i v i d and baseline percentages, revealed that the VM d i s t r i b u t i o n was, i n fact, s i g n i f i c a n t l y r e l a t e d to the baseline d i s t r i b u t i o n (r = .65, p_ < .01). Inspection of Figure 1 shows l i t t l e difference between v i v i d and baseline memories for 15 of the 19 categories. This was confirmed by an i n d i v i d u a l chi-square analyses that showed no s i g n i f i c a n t differences for these 15 V i v i d Autobiographical Memories 64 categories. This finding suggests that, for most of the thematic categories, the number of VMs reported i s representative of the number of events that subjects have av a i l a b l e i n memory. Of the four categories that were found to be s i g n i f i c a n t l y d i f f e r e n t , two categories were found to be s i g n i f i c a n t l y overrepresented i n the VM sample compared to baseline percentages. These categories were accidents, i n j u r i e s , i l l n e s s e s , and operations (chi-square = 102.2, p_ < .001), and unpleasant childhood landmarks (chi-square = 12.4, p_ < .001). The remaining two categories vacations (chi-square = 23.41, p. < .001) and pleasant childhood landmarks ( c h i -square = 11.01, p_ < .001) were found to be underrepresented i n the VM sample compared to baseline percentages. These findings indicate that the frequency with which VMs from each of these four categories were reported, does not appear simply to r e f l e c t the r e l a t i v e frequency of r e a d i l y a v a i l a b l e memories that the subjects possess. On an i n t u i t i v e l e v e l , several c h a r a c t e r i s t i c s that are usually associated with the overrepresented categories and not with the underrepresented c h a r a c t e r i s t i c s , seem possible, including unpleasant a f f e c t , high arousal, and consequentiality. A f f e c t and v i v i d memories. A t h i r d important finding i s that the VMs produced i n Phase 1 were almost three times more l i k e l y to be unpleasant then pleasant events. F i f t y two percent of the VMs were for unpleasant events and only 19% were pleasant. The remaining 3 0% of memories were c l a s s i f i e d V i v i d Autobiographical Memories 65 as eith e r ambiguous (9%) or as neutral (20%). A complete theory of VMs should illuminate the reasons f o r t h i s bias toward unpleasant events. To illuminate the differences between pleasant and unpleasant memories, a number of t - t e s t s were conducted (see Table 1). Although unpleasant memories were found to be s i g n i f i c a n t l y older then pleasant memories, they were not rated s i g n i f i c a n t l y higher on the expected enduring e f f e c t (EE) scale. Further, unpleasant memories were found to be rated lower on the actual e f f e c t (AE) scale and to be rehearsed l e s s often than pleasant memories. The s p e c i a l mechanism hypothesis predicts that, on average, more VMs should be rated higher for EE than AE. On the other hand, Neisser's view predicts that actual consequences (AE) and rehearsal should be higher f o r VMs than for non-VMs. I f unpleasant events are more l i k e l y to produce VMs than pleasant events, the same pattern predicted to e x i s t between VMs and non-VMs should be found between pleasant and unpleasant memories. Neither predicted r e s u l t s were found. Unpleasant memories were found to be rated higher for EE than AE. Further, unpleasant memories were not found to be rated higher than pleasant memories for AE or rehearsal. Therefore, these r e s u l t s support neither the spe c i a l mechanism theory, nor Neisser's view. The high percentage of unpleasant events reported i n t h i s study does not necessarily mean that VMs are more l i k e l y the r e s u l t of unpleasant events. The high V i v i d Autobiographical Memories 66 percentage may be a t t r i b u t a b l e to methodological factors. This w i l l be discussed more f u l l y at the end of t h i s section. Necessity of Consequentiality. The fourth finding from Experiment 1 concerns the r e l a t i o n s h i p between consequentiality and VMs. Brown and Kulik's s p e c i a l mechanism theory p r e d i c t s that EE ratings would, on average, be higher than AE ratings. This was not found to be true. The mean differe n c e score between EE and AE was -0.3 (range +6 to -6). In other words, AE was rated higher than EE. Further, v i v i d memories existed f o r events that were not expected to be consequential, but l a t e r turned out to be. Therefore, i t does not appear that expected consequentiality i s necessary f o r VMs as Brown and Kulik suggested. Consequentiality and Rehearsal. A further r e s u l t of i n t e r e s t i n t h i s study, concerns the r e l a t i o n s h i p between consequentiality and rehearsal. Correlations calculated between EE and rehearsal, and AE and rehearsal, were + .24 and +.55 respectively, i n d i c a t i n g that memories that are rated to have higher consequentiality ratings, are also thought about more often. This suggests that the r e l a t i o n s h i p between consequentiality and VMs as proposed by Brown and Kulik might be explained by the r o l e that rehearsal plays i n maintaining the memory. Persistence and V i v i d Memories. The in s t r u c t i o n s directed subjects to report only memories for events that occurred more than two years ago. This r e s t r i c t i o n was V i v i d Autobiographical Memories 67 included because VMs are only remarkable i f they are also p e r s i s t e n t . Subjects were, therefore, presented with two search tasks; finding v i v i d memories and f i n d i n g memories that occurred more than two years ago. I f subjects conducted t h i s search by f i r s t asking themselves what they were doing two years ago, then they would be biased towards s e l e c t i n g v i v i d memories from that time period, and the memory age d i s t r i b u t i o n would be to be skewed toward two year old memories. I f they searched f o r v i v i d memories, then checked to see i f memories that came to mind passed the two year or older c r i t e r i o n , then no such skewing would be expected. The mean rated age of the reported memories was 7.5 years (mode = 3.5 years, median = 5 years, range = 2 to 34 years). Therefore, i t appears that the subjects viewed searching for v i v i d memories to be t h e i r primary task. Discussion The main finding from Experiment 1—the frequency of VMs found i n each thematic c a t e g o r y — i s consistent with the r e s u l t s of previous studies. For example, for the category of accidents, i n j u r i e s , i l l n e s s e s , and operations, our findings are remarkably consistent with those of Rubin and Kozin (1984) who found that 17.8% of t h e i r subjects' VM reports were for accidents or i n j u r i e s . The categories i n Experiment 1 were more narrowly defined than i n Rubin and Kozin's. I f memories for witnessing accidents and l i f e threatening events were V i v i d Autobiographical Memories 68 included i n the accident, injury, i l l n e s s , and operations category i n t h i s study, as they were i n Rubin and Kozin's study, the present studies percentage for t h i s category would be 16.6%, which i s comparable to the r e s u l t s found by Rubin and Kozin. The present findings are also s i m i l a r to those of Cohen and Faulkner (1988), whose combined percentage of VMs for p h ysical trauma and i l l n e s s e s was (19.7%). Therefore, the thematic breakdown of v i v i d memories found i n t h i s study appears to capture the same general a t t r i b u t e s of VMs. The s i m i l a r i t y i n r e s u l t s across studies, provides evidence that the categorization scheme arrived at i n t h i s study, r e f l e c t s stable thematic c h a r a c t e r i s t i c s of VMs. This consistency between the categorization percentages across studies i s p a r t i c u l a r l y remarkable given the differences (geographic, age range, cohort groups) that existed between the three subject groups. Rubin and Kozin's, and Cohen and Faulkner's subjects cannot be expected to have experienced the same types of l i f e events with the same frequency as the subjects used i n Experiment 1. This i s p a r t i c u l a r l y true given the age range of the subjects used i n Cohen and Faulkner's study. The l i f e experiences of a subject population, such as the one studied by Cohen and Faulkner, which includes members from such a wide range of age cohorts, cannot be expected to be s i m i l a r to the l i f e experiences of u n i v e r s i t y student populations. Although the thematic categorization percentages were not expected to generalize to V i v i d Autobiographical Memories 69 other subject groups, the comparability of the r e s u l t s from these three studies supports the second finding of t h i s study--that the frequency with which some events are reported as v i v i d , does not simply r e f l e c t the underlying frequency of baseline memories the subjects possess. I t i s also i n t e r e s t i n g to note that there were no s i g n i f i c a n t differences found for f i f t e e n of the nineteen categories. This suggests that there i s a great deal of s i m i l a r i t y between what subjects report as VMs and the r e a d i l y a v a i l a b l e memories used as a baseline measure i n t h i s study. In f a c t , i t would be l o g i c a l to assume that, when enumerating r e a d i l y a v a i l a b l e memories, subjects also enumerated VMs. In other words, VMs are a sub-population of the baseline memories. Therefore, the differences found between these two groups could be underestimated. I f a control group of non-v i v i d memories were used, t h i s bias could be eliminated. This issue w i l l be addressed more f u l l y i n Experiment 2. In comparing the thematic content of VMs to baseline memories, the most s t r i k i n g difference i s the s i g n i f i c a n t overrepresentation of accidents, i n j u r i e s , and i l l n e s s e s and of unpleasant childhood landmarks among the VMs. In contrast to t h i s over-representation of r e l a t i v e l y unpleasant events, the two s i g n i f i c a n t l y under-represented categories among the VMs, are vacations and pleasant childhood landmarks, both of which are r e l a t i v e l y pleasant events. I t i s , therefore, suggested from the findings i n t h i s study, that c e r t a i n V i v i d Autobiographical Memories 70 categories of unpleasant events are more l i k e l y to be v i v i d l y r e c a l l e d , whereas the reverse i s true for c e r t a i n pleasant events. Of the 444 memories that could be c l e a r l y c l a s s i f i e d as a f f e c t i v e l y pleasant or unpleasant, the overwhelming majority (326) were unpleasant. Consistent with the thematic content findings, which showed unpleasant childhood events to be over-represented i n the VM sample, unpleasant VM events occurred at a younger age than pleasant VM events. There are two possible explanations f o r t h i s . F i r s t , i f the sp e c i a l mechanism theory i s true and VMs are permanent, then the younger age of encoding found f o r unpleasant memories may r e f l e c t the f a c t that such unpleasant events are more l i k e l y to be VMs. On the other hand, ordinary memory theories suggest that factors such as rehearsal might explain t h i s difference i n memory age. I t could be that we think about unpleasant events more so than pleasant events and, thus, maintain the memory representation v i v i d l y i n memory over a longer period of time. The fa c t that unpleasant VMs were rehearsed less often than pleasant memories, however, makes the overwhelming proportion of negative events reported d i f f i c u l t to explain i n terms of rehearsal. A second possible explanation i s that unpleasant events are more arousing and, thus, are better encoded at the time they occur. This may lead to t h e i r being retained over a longer period of time. This issue w i l l be more c a r e f u l l y explored i n Experiment 2. V i v i d Autobiographical Memories 71 A further finding of importance i n t h i s study, i s that, counter to Brown and Kulik*s theory, high expected consequentiality i s not necessary for the existence of VMs. The f a c t that the c o r r e l a t i o n between AE and rehearsal i s s i g n i f i c a n t l y higher than between EE and rehearsal (p_ < .01) further supports Neisser's view that any r e l a t i o n s h i p between consequentiality and v i v i d r e c a l l , may be explainable i n terms of the influence rehearsal has on maintaining the memory representation. Experiment 2 w i l l examine the r e l a t i o n s h i p s among vividness, rehearsal, consequentiality, and a f f e c t more f u l l y . I t i s possible, of course, that the thematic content differences between the Phase 1 and Phase 2 memories are not due to vividness, but are, instead, a product of the d i f f e r e n t procedures used to enumerate two memory groups. Therefore, a more i d e a l control group would be one that was c o l l e c t e d i n the same manner as the VMs. Such a control group was employed i n Experiment 2. Experiment Two The main focus of the second study was to explore the thematic and a f f e c t i v e c h a r a c t e r i s t i c s of v i v i d and non-vivid memories, as well as the relationships among consequentiality, a f f e c t , arousal, rehearsal, and vividness. The following questions were examined. (a) Are some l i f e - e v e n t s more l i k e l y to produce v i v i d memories than others? (b) What i s the V i v i d Autobiographical Memories 72 r e l a t i o n s h i p between consequentiality and vividness, and can t h i s r e l a t i o n s h i p be explained by rehearsal? (c) What are the r e l a t i o n s h i p s between d i r e c t i o n of a f f e c t and vividness, and between arousal and vividness? To examine these questions, subjects were asked to write an account of an event that they had personally experienced. The event was required to have taken place two or more years ago, but i t was not required to be v i v i d . Thus, persistent memories of varying degrees of vividness were sampled. A f t e r w r i t i n g the account, subjects were asked to rate t h e i r memory of the reported event on a vividness scale as well as several other scales. This method has two important advantages. The f i r s t i s that VMs and non-VMs were e l i c i t e d using the same procedure. The second advantage i s that c o r r e l a t i o n a l analyses could be performed, thereby enabling an examination of the nature and degree of relationships among vividness, consequentiality, arousal, pleasantness, and rehearsal. The main difference between the manner i n which Experiment 1 explored s i m i l a r questions i s that Experiment 2 assessed vividness as a continuous variable, whereas Experiment 1 treated vividness as though i t were a d i s c r e t e v a r i a b l e . To examine the thematic contents of memories, subjects i n Experiment 2 were provided with the category names derived from Experiment 1 and were asked to s e l e c t the category into which t h e i r own memory protocol best f i t . In the i n t e r e s t of c l a r i t y , two minor changes were made to the thematic V i v i d Autobiographical Memories 73 categories (see Appendix B). F i r s t , because any influence of memory age could be assessed d i r e c t l y i n Experiment 2, d i f f e r e n t i a t i o n s between adult and childhood categories were not included. Second, the categories of f i r s t hearing about deaths, and f i r s t hearing bad news, which were combined i n Experiment 2, were l i s t e d as separate categories. A t o t a l of twenty thematic categories were used i n Experiment 2. Experiment 2 included separate scales to assess vividness, consequentiality and rehearsal. As i n Experiment 1, consequentiality was operationally defined i n two ways, as expected enduring e f f e c t (EE), and as actual e f f e c t (AE). The s p e c i a l mechanism theory predicts that EE should be more c l o s e l y r e l a t e d to vividness than AE because only factors that are present at the time of the event could serve to t r i g g e r the \"Now P r i n t \" mechanism. On the other hand, ordinary memory theories do not require a d i r e c t r e l a t i o n s h i p between vividness and consequentiality. Consequential events may, however, be thought about more often than events that are inconsequential, and any r e l a t i o n s h i p between actual consequentiality and vividness may be explainable i n terms of rehearsal. Experiment 2 also examined the r o l e of a f f e c t i n producing VMs. In order to assess various aspects of a f f e c t , subjects were asked to complete two Mehrabian and Russell (1974) a f f e c t scales. The f i r s t scale was to be completed according to a f f e c t experienced when the event occurred, and V i v i d Autobiographical Memories 74 the second was completed according to a f f e c t at the time of r e c a l l . The Mehrabian and Russell scales assess two orthogonal dimensions of af f e c t — p l e a s a n t n e s s and arousal. Scores on the pleasantness dimension can range from -24 for extremely unpleasant events to +24 for extremely pleasant events. Absolute ratings along t h i s dimension can be used to assess the degree or i n t e n s i t y of a f f e c t independent of i t s d i r e c t i o n . Brown and Kulik proposed that some c r i t i c a l l e v e l of \"emotional arousal\" may t r i g g e r the sp e c i a l mechanism. However, i t i s unclear what i s meant by emotional arousal. Experiment 2 operationalized t h i s term i n two ways, as arousal and as absolute pleasantness. Ordinary memory theories would pre d i c t that both of these factor may be associated with vividness; they are factors that can boost o v e r a l l memory performance. I t i s possible that events that are arousing are be better attended to and better encoded. Further, events that involve extreme absolute values of pleasantness are more l i k e l y to be a f f e c t i v e l y d i s t i n c t i v e , thereby also conferring a memory advantage. Although both the spe c i a l mechanism and ordinary memory theories predict a p o s i t i v e r e l a t i o n s h i p between a f f e c t i v e factors and vividness, the natures of the predicted r e l a t i o n s h i p s d i f f e r . For the sp e c i a l mechanism view, e i t h e r factor could t r i g g e r the \"Now P r i n t \" mechanism provided i t was present at a c r i t i c a l l y high l e v e l at the time the event occurred. Also, because a f f e c t i v e factors are part V i v i d Autobiographical Memories 75 of the perceptual c h a r a c t e r i s t i c s that are thought to be permanently encoded and re-experienced upon r e c o l l e c t i o n , there should be l i t t l e change, i f any, between ratings of a f f e c t at the time the event occurred and when i t i s r e c a l l e d . A f i n a l methodological difference between Experiment 1 and 2 involved the number of memories e l i c i t e d per subject. In Experiment 1, the requirement to produce and then rate each memory protocol may have influenced what subjects produced as the second and t h i r d memory. For t h i s reason, subjects i n Experiment 2 were required to r e c a l l only one memory, and further, they wrote the protocol before turning the page of the t e s t booklet and looking at the r a t i n g scales. Method Two hundred f i f t y nine introductory psychology students (mean age 19.7 years) p a r t i c i p a t e d for course c r e d i t . Of these, 126 completed the questionnaire during c l a s s time. The other 133 subjects received a take-home version of the same questionnaire (see Appendix B for complete questionnaire). The subjects were given a booklet which consisted of four pages. On the f i r s t page, subjects were instructed to describe b r i e f l y a personal episode (something i n which they themselves were involved) that occurred two or more years ago. On the second page of the booklet, subjects were asked to write t h e i r current age and an estimate of t h e i r age at the time the event took place. Further, the second and t h i r d page V i v i d Autobiographical Memories 76 of the booklet contained scales to assess the following memory q u a l i t i e s ; vividness, rehearsal, expected enduring e f f e c t (EE), actual e f f e c t (AE), i n t e n s i t y of a f f e c t experienced at the time of the event (INTT) and at r e c a l l (INTN), the percentage of the emotions experienced which were pleasant at the time the event occurred (%PLT) and at r e c a l l (%PLN), and two Mehrabian and Russell (1974) a f f e c t scales. Subjects completed the f i r s t a f f e c t scale according to t h e i r feelings at the time the event took place and the second according t h e i r f e e l i n g s upon r e c a l l i n g the event. The t e s t booklet i s presented i n Appendix C. The f i n a l page of the booklet contained a l i s t of twenty thematic categories from Experiment 1 and subjects were instructed to indicate into which category t h e i r r e c a l l e d event best f i t . Subjects who completed the questionnaire i n c l a s s were allowed up to h a l f an hour. Those subjects who completed the booklets at home were simply required to return i t within one week of receiving i t . Results The primary purpose of Experiment 2 was to compare the thematic c h a r a c t e r i s t i c of VMs and non-VMs. The f i r s t step, therefore, was to operationally define the terms VM and non-VM within t h i s study. The memories c o l l e c t e d ranged i n vividness ratings from 2 to 7. The mean vividness r a t i n g was 5.3. Forty-six percent of the memories received vividness ratings of e i t h e r 6 or 7. An approximate median s p l i t was used to V i v i d Autobiographical Memories 77 categorize the memories as eithe r v i v i d or non-vivid. Memories defined as v i v i d were rated 6 or 7 (n = 119, mean = 6.2), and non-vivid memories were defined as having ratings of less than 6 (n = 140, mean = 4.3). T-test analyses revealed no s i g n i f i c a n t differences between the take-home and i n - c l a s s subjects groups, and thus the r e s u l t s from the two groups were compiled f o r a l l subsequent analysis. Thematic differences between v i v i d and non-vivid memories. The most important finding i n t h i s study i s the percentages of VM and non-VM memories that f e l l into each of the thematic categories (see Figure 2). Less then two percent of a l l the memory protocols f e l l into each of the four following categories: b i r t h s , relocating, car accidents with no i n j u r i e s , and nature related events. For ease of analysis and presentation, memory protocols for these events were placed into the appropriate \"pleasant other\", or \"unpleasant other\" category based upon t h e i r Mehrabian and Russell ratings. This l e f t sixteen categories that were used i n subsequent analyses. Similar to the r e s u l t s found between v i v i d reported memories and estimated a v a i l a b l e memories i n Experiment 1, the r e s u l t s showed l i t t l e difference between the percentages of VMs and non-VMs for the majority (13/16) of the categories. An o v e r a l l Pearson c o r r e l a t i o n showed that s i m i l a r percentages of VMs and non-VMs were produced across categories (r = .65). Nevertheless, i n d i v i d u a l chi-square analyses revealed s i g n i f i c a n t differences for three of the 16 V i v i d Autobiographical Memories 78 categories. Both l i f e threatening events (chi-square = 35.1, p_ < .001), and accidents, i n j u r i e s , i l l n e s s e s , or operations (chi-square = 11.51, p_ < .001) were found to occur more frequently i n the VM sample than i n the non-vivid sample. In contrast, more vacation memories were found i n the non-VM than the VM category (chi-square = 7.56, p_ < .01). The f i r s t f i n d i n g f i t s with the spe c i a l mechanism view. By t h i s view the a b i l i t y to form v i v i d , persistent memories fo r l i f e -threatening and injurious events has s u r v i v a l value. As Pettigrew (1978) points out: It i s easy to see that i t would have made l i t t l e sense f o r one of the brains of our predecessors to have gone on endlessly memorizing the d e t a i l s of contours of the c e i l i n g of hi s l a i r i f he were l y i n g there safe, sound and satiated; but that i t would have made sense to record even apparently 1 t r i v i a l 1 sensory d e t a i l s associated with a l i f e - t h r e a t e n i n g encounter (p. 73). Although the data from both Experiment 1 and Experiment 2 support the idea that l i f e threatening or injurious events are more l i k e l y to be v i v i d l y r e c a l l e d , no spe c i a l mechanism i s required f o r explaining t h i s f i nding. I t could be that ordinary memory processes work i n such a way as to produce the same type of s u r v i v a l value that Pettigrew a t t r i b u t e s to a spe c i a l mechanism. In other words, i t could be that l i f e threatening events are high on factors that boost memory performance i n general. The second finding, that vacation memories are less l i k e l y to be v i v i d l y than non-vividly r e c a l l e d , i l l u s t r a t e s V i v i d Autobiographical Memories 79 the necessity of using a control group to study thematic content. Knowing the percentage of VMs which f a l l into any thematic group does not necessarily t e l l us anything about VMs or what might give r i s e to them. The percentages may simply r e f l e c t the number of events i n each category which subjects have experienced and remember experiencing. For example, Cohen and Faulkner (1988) noted that f i v e percent of the memory protocols from t h e i r study could be c l a s s i f i e d as vacation memories. This percentage i s i d e n t i c a l to the percentage of v i v i d vacation memories reported i n Experiment 2. Whereas, Cohen and Faulkner's study suggested that vacation memories were a representative c l a s s of VMs, Experiment 2 revealed that they are, i n fact, much more representative of non-VMs. Consequentiality and Vividness. The second area of concern i n t h i s study was the r e l a t i o n s h i p between vividness and consequentiality. T-test analyses showed that VMs were rated as s i g n i f i c a n t l y higher for both EE and AE i n d i c a t i n g that VMs are perceived to be more consequential both at the time they occur and at the time of r e c o l l e c t i o n (see Table 2). Both EE and AE were s i g n i f i c a n t l y correlated with vividness (see Table 3). As i n Experiment 1, a wide range of consequentiality ratings (1 to 7) were found for VMs i n d i c a t i n g that consequentiality i s not necessary for v i v i d r e c a l l . Inspection of a s c a t t e r p l o t confirmed that the r e l a t i o n s h i p between consequentiality and vividness i s V i v i d Autobiographical Memories 80 continuous. Therefore, consequentiality cannot be considered a t r i g g e r f o r the sp e c i a l mechanism. Ordinary memory theories predict that the r e l a t i o n s h i p between consequentiality and vividness might be explained i n terms of rehearsal. As suggested by the r e s u l t s of Experiment 1, consequential events may be thought about more often than non-consequential events, and, therefore, may be better retained than non-consequential events. Rehearsal was found to have the strongest c o r r e l a t i o n with vividness, (r = .40). In order to explore rehearsal's r o l e i n maintaining memory vividness, a p a r t i a l c o r r e l a t i o n was conducted c o n t r o l l i n g for rehearsal (see Table 4 ) . I t was found that when the e f f e c t s of rehearsal were p a r t i a l l e d out, the c o r r e l a t i o n s between EE and vividness, and between AE and vividness were no longer s i g n i f i c a n t . Conversely, i f EE and AC were c o n t r o l l e d for, the c o r r e l a t i o n between rehearsal and vividness remained s i g n i f i c a n t (r = .31). Therefore, vividness of r e c a l l cannot be explained i n terms of consequentiality. Instead the rel a t i o n s h i p between consequentiality and vividness i s explainable by the re l a t i o n s h i p between consequentiality and rehearsal. A f f e c t and Vividness. The t h i r d concern of Experiment 2 was the r e l a t i o n s h i p between a f f e c t and vividness. Although VMs tended to be rated as less pleasant than non-vivid memories, t h i s difference was not s i g n i f i c a n t (see Table 2). Overall, 48.3% of the memories reported i n Experiment 2 were V i v i d Autobiographical Memories 81 rated as unpleasant at the time they occurred, 50.2% were rated as pleasant, and the remaining 0.8% were rated as neutral. This finding suggests that the over-reporting of unpleasant VMs i n Experiment 1 was not due to a general tendency to report unpleasant events nor does i t r e f l e c t a tendency to r e c o l l e c t unpleasant events i n a v i v i d manner. Instead, t h i s difference was more l i k e l y due to the categorization procedure used. Subjects' percentage ratings of pleasantness (%PLT and %PLN) and the Mehrabian and Russell (1974) scores (PLT and PLN) were found to correlate .85 and .96, respectively, for ratings at the time the event occurred and at r e c a l l . In view of these high co r r e l a t i o n s , i t was assumed that the percentage a f f e c t ratings and the Mehrabian and Russell scales index the same dimensions, and, thus, to simplify further analyses, only the Mehrabian and Russell scales were used. T-test analyses (see Table 2) showed that VMs were s i g n i f i c a n t l y more arousing both at the time the event occurred and at r e c a l l . Further, both ART and ARN were found to be s i g n i f i c a n t l y correlated with vividness (see Table 3). A wide range of ART ratings (+24 to -18) were found f o r VMs. Inspection of a s c a t t e r p l o t indicated that the re l a t i o n s h i p between ART and vividness was continuous, thereby suggesting that the r e l a t i o n s h i p between arousal and vividness i s not of the discontinuous nature that i s required for arousal to serve as a t r i g g e r for a special mechanism. V i v i d Autobiographical Memories 82 No s i g n i f i c a n t c o r r e l a t i o n was found between absolute pleasantness then (absPLT) ratings and vividness, thereby <, suggesting that absPLT i s also not a t r i g g e r f o r a s p e c i a l memory mechanism. Vividness was, however, s i g n i f i c a n t l y associated with absolute pleasantness now (absPLN). This reinforces the idea that pleasantness (or unpleasantness) i s one of the perceptual q u a l i t i e s re-experienced upon production of a VM. The f a c t that absPLN i s s i g n i f i c a n t l y correlated with rehearsal (r = .30), whereas p r a c t i c a l l y no c o r r e l a t i o n was found between absolute PLT and rehearsal (r = .10), indicates that any c o r r e l a t i o n between absolute pleasantness and vividness can be explained i n terms of rehearsal. S i g n i f i c a n t c orrelations were found between the i n t e n s i t y v a r i a b l e s (INTT and INTN) and vividness (see Table 3). In including the i n t e n s i t y scales, i t was thought that subjects would t r e a t them ei t h e r as arousal or as absolute pleasantness. The correlations between the i n t e n s i t y v a r i a b l e s and absolute pleasantness variables (absPLT and absPLN), and between i n t e n s i t y and arousal variables (ART and ARN) were found to be s u r p r i s i n g l y low (see Table 4). P a r t i a l c o r r e l a t i o n s were conducted between the i n t e n s i t y variables (INTT and INTN) and vividness with the time appropriate arousal and absolute pleasantness variables c o n t r o l l e d for. The r e s u l t i n g p a r t i a l correlations between vividness and INTT, and between vividness and INTN were .19 and .30, respectively. Therefore, whatever subjects were in t e r p r e t i n g i n t e n s i t y to V i v i d Autobiographical Memories 83 mean, i t i s neither a r a t i n g of degree of absolute pleasantness, nor i s i t general arousal. A possible explanation i s that i n t e n s i t y , as measured i n t h i s experiment, was interpreted by the subjects as being somewhat synonymous with vividness. I f t h i s i s the case, the s i g n i f i c a n t c o r r e l a t i o n s found between i n t e n s i t y and vividness are e a s i l y explained. Persistence and Vividness. Unlike Experiment 1, i n which the age-range of memories was large (2 to 42 years), the age-range of memories found i n Experiment 2 was much smaller (2 to 17). Eighty percent of the memories produced i n Experiment 2 were between 2 and 5 years old. One p o s s i b i l i t y that might explain the difference i n memory age-range between experiments i s that i n the f i r s t experiment, subjects were searching for a \" v i v i d memory\" as t h e i r primary task, whereas, i n the second experiment, subjects may have searched t h e i r memories by asking, \"What was I doing two years ago?\" This question may have been suggested by the instructions to describe an event which took place two years or more ago. VMs and non-VMs were of s i m i l a r age (mean memory age, VMs = 5.5, Non-VMs = 6.2) and there was no s i g n i f i c a n t c o r r e l a t i o n between age of memory and vividness. Discussion Two i n t e r e s t i n g conclusions are suggested by the r e s u l t s from the content analysis. F i r s t , the r e s u l t s showed V i v i d Autobiographical Memories 84 systematic and predictable differences i n the vividness with which d i f f e r e n t l i f e events were r e c a l l e d . Second, they showed that whatever factor(s) underlie vividness, these factors were more l i k e l y to be associated with memories of l i f e - t h r e a t e n i n g and injurious events (accidents, i n j u r i e s , i l l n e s s e s , and operations), and less l i k e l y to be associated with vacation memories. The o v e r a l l r e s u l t s from t h i s study provide no support for Brown and Kulik's s p e c i a l mechanism theory. Although, as predicted by the s p e c i a l mechanism view, the r e s u l t s showed a r e l a t i o n s h i p between vividness and consequentiality, t h i s r e l a t i o n s h i p was explainable primarily by rehearsal. Further, the nature of the relationships between vividness and consequentiality, a f f e c t and rehearsal did not f i t the pattern that would be expected according to the s p e c i a l mechanism theory. For consequentiality, absPL, and arousal, the \"now\" v a r i a b l e s (AE, absPLN, and ARN) were better predictors of vividness than the \"then\" variables (EE, ART, and absPLT). Therefore, counter to what i s implied by Brown and Kulik's FB theory, the encoding of VMs does not appear to be the product of a s p e c i a l memory mechanism triggered by factors that are present at the occurrence of the event. F a i l u r e to f i n d support for Brown and Kulik's model does not mean that VMs can be explained within ordinary memory theories. Experiment 2 has found that two factors, rehearsal and arousal, that are generally associated with superior V i v i d Autobiographical Memories 85 memory performance within ordinary memory theories, are also r e l a t e d to vividness. While t h i s suggests that ordinary memory theories may explain the persistence and d e t a i l of r e c a l l associated with VMs, i t does not explain why some persistent and d e t a i l e d memories are also r e c a l l e d with perceptual c l a r i t y while others are not. Cues as to what factors might d i f f e r e n t i a t e between v i v i d memorable events and non-vivid memorable events are suggested by the thematic content analyses. The estimated percentages of a v a i l a b l e memories (from Experiment 1) showed that subjects tended to have several memories for vacations and r e l a t i v e l y few for l i f e - t h r e a t e n i n g and injurious events. Therefore, v i v i d l y r e c a l l e d events appear to be more d i s t i n c t i v e than non-vividly r e c a l l e d events. The influence of d i s t i n c t i v e n e s s w i l l be examined i n Experiment 3. Experiment Three The main focus of the t h i r d experiment was to more c l e a r l y define and assess the r e l a t i o n s h i p between vividness and factors that are generally held to be associated with superior memory performance. In addition to addressing the r o l e s of arousal, rehearsal, consequentiality, and a f f e c t as i n Experiment 2, Experiment 3 included several scales to examine the r e l a t i o n s h i p between various types of d i s t i n c t i v e n e s s and vividness. The following questions were examined. (a) What i s the r e l a t i o n s h i p between d i f f e r e n t V i v i d Autobiographical Memories 86 aspects of d i s t i n c t i v e n e s s and vividness? (b) Can factors that are generally associated with superior memory performance (arousal, rehearsal, and distinctiveness) explain the existence of VMs? (c) Does the r e l a t i o n s h i p between expected enduring e f f e c t and vividness, as found i n Experiment 2, generalize to other d e f i n i t i o n s of consequentiality? (d) Are VMs more l i k e l y to be associated with unpleasant than pleasant events? In order to examine these questions, the general method used i n Experiment 2 was employed i n Experiment 3. Subjects were asked to write an account of an event that they had personally experienced, with the constraint that the event occurred two or more years ago. Unlike Experiment 2, subjects were provided with the t e s t booklet only a f t e r they had f i n i s h e d w r i t i n g t h e i r memory accounts. The subjects completed the t e s t booklet by r a t i n g t h e i r memory and the event associated with i t on a v a r i e t y of scales, including d i s t i n c t i v e n e s s , a f f e c t , rehearsal, consequentiality, and vividness (see Appendix C). The main new issue examined i n Experiment 3, was the r e l a t i o n s h i p between d i s t i n c t i v e n e s s and vividness. Four separate scales were used to assess d i f f e r e n t aspects of d i s t i n c t i v e n e s s . The f i r s t scale required r a t i n g how d i s t i n c t i v e the reported event was i n r e l a t i o n s h i p to other events the subject had experienced (dSELF). The second scale required r a t i n g d i s t i n c t i v e n e s s r e l a t i v e to events the subject V i v i d Autobiographical Memories 87 perceived others to have experienced (dOTHER). A t h i r d set of scales was included to assess the di s t i n c t i v e n e s s of the loc a t i o n (dLOC), the a c t i v i t i e s (dACT), the people involved (dPPL), the emotions experienced (dEM), and the thoughts (dTHT) that were going through the subject's mind when the event occurred. F i n a l l y , i n order to assess temporal d i s t i n c t i v e n e s s , subjects were asked to indicate whether the event they reported was best described as an \"only time\", \" f i r s t time\", \" l a s t time\", or \"one of a s e r i e s \" event. The exact wordings used to define each scale i s i n Appendix C. Ordinary memory theories propose that any c h a r a c t e r i s t i c ( s ) which d i f f e r e n t i a t e s an event from s i m i l a r ones, reduces the a b i l i t y of the memory system to assimilate the event into e x i s t i n g schemata. Storage of memories i n schematic form i s economical and integrative, but i t produces confusions between memories for in d i v i d u a l events. Therefore, events that are d i s t i n c t i v e are more l i k e l y to be re t r i e v a b l e as i n d i v i d u a l instances. D i s t i n c t i v e aspects of an event may serve as cues to aid r e c a l l for additional d e t a i l s of the o r i g i n a l event. In view of the central r o l e of di s t i n c t i v e n e s s i n remembering, i t i s possible that ei t h e r an extreme l e v e l of di s t i n c t i v e n e s s , or perhaps a s p e c i f i c aspect of d i s t i n c t i v e n e s s underlies the existence of VMs. This area has never been empirically examined. The second issue of concern i n Experiment 3 was whether or not the p o s i t i v e r e l a t i o n s h i p between vividness and V i v i d Autobiographical Memories 88 enduring e f f e c t found i n Experiment 2, can be generalized to other d e f i n i t i o n s of consequentiality that have been used i n previous studies. Consequentiality was assessed i n two ways i n addition to EE and AE. F i r s t , by means of two scales that were labeled expected and actual consequentiality (EC and AC, respectively) and second by means of two scales that were labeled expected and actual importance (EI and AI, r e s p e c t i v e l y ) . I t i s possible that the EE and AE scales from Experiments 1 and 2 did not assess the same concept that previous researchers have referred to as consequentiality (Brown & Kulik, 1977; Rubin & Kozin, 1984; Pillemer, 1984) or as importance ( Rubin & Kozin, 1984; Cohen & Faulkner, 1988; Conway & Berkerian, 1988). This concern i s p a r t i c u l a r l y relevant i n l i g h t of Rubin and Kozin's finding that \"personal importance\" and \"consequentiality\" ratings used i n t h e i r study correlated only .44 with each other. Using d i f f e r e n t d e f i n i t i o n s , i t might be found that consequentiality i s r e l a t e d to vividness i n the discontinuous, necessary manner required by Brown and Kulik's model. The t h i r d issue of concern i n t h i s experiment was the degree to which factors, normally associated with memory quality, can be used to explain VMs. Rehearsal and arousal (ART) were defined i n Experiment 3 i n the same manner as they were defined i n Experiment 2. I f the factors generally thought to underlie superior memory performance (arousal, rehearsal, and distinctiveness) can be used to explain a V i v i d Autobiographical Memories 89 s i g n i f i c a n t amount of the variance found i n vividness ratings, then the postulation of a spe c i a l memory mechanism theory would be unparsimonious. The fourth issue examined i n Experiment 3 i s the re l a t i o n s h i p between pleasantness and vividness. As i n Experiment 2, pleasantness was operationally defined i n the Experiment 3 by ratings on the Mehrabian and Russell a f f e c t scales (PLT). In Experiment 1, i t was found that the majority of a f f e c t i v e l y c l a s s i f i a b l e memories were unpleasant. In Experiment 2, i t was found that unpleasant memories were neither more l i k e l y to be reported than pleasant memories, nor were they found to be s i g n i f i c a n t l y correlated with vividness. I t i s unclear why such an overwhelming number of memories f o r unpleasant events were reported i n Experiment 1 and, therefore, t h i s issue merits further in v e s t i g a t i o n . Method Two, hundred and three introductory psychology students (mean age = 19.7) pa r t i c i p a t e d for course c r e d i t . Unlike Experiments 1 and 2, these subjects were tested i n d i v i d u a l l y or i n small groups. Upon entering the t e s t i n g room, subjects were given an Information/Consent sheet. This sheet outlined the purpose of the experiment as follows: We are simply interested i n having you b r i e f l y t e l l us about one event from your l i f e and then to rate your memory for t h i s event on a series of scales. From t h i s we hope to learn something about the types of events people remember, t h e i r reactions to these events as well as about how such events are represented i n memory. V i v i d Autobiographical Memories 90 Aft e r having read the Information/Consent form, subjects were given a s i n g l e sheet of paper with the following i n s t r u c t i o n s : In the space provided on t h i s page, please describe b r i e f l y a personal memory of an event (something i n which you yourself were involved) that occurred two years ago or longer. The event should be of a short duration covering no more than one day. They were allowed as much time as they needed to write t h e i r memory protocols. Upon completion, the memory protocol sheet was taken away from the subject. This ensured that the subjects had selected and written t h e i r memory accounts before seeing the scales on which they were to rate them. They were then given a four page t e s t booklet containing 21 scales (see Appendix C). Unlike Experiment 2 where both seven-point numbered and 9-point unnumbered scales were used (see Appendix B), a l l scales i n t h i s study were of the nine-point, unnumbered format used by Mehrabian and Russell. The areas assessed by these scales were: vividness, covert and overt rehearsal (RC and RO), consequentiality, (EE, AE, EI, AI, EC, and AC), di s t i n c t i v e n e s s (dSELF, dOTHER, dLOC, dACT, dPPL, dEM, and dTHT), a f f e c t , (ART, and PLT), confidence (CON), surprise (SUR), l i k e l i h o o d of the event occurring (LE), and scales to assess the degree of change t h i s event had on the manner i n which subjects thought about themselves, other people or the world i n general (see Appendix C for complete l i s t i n g ) . Six versions of the r a t i n g scale booklets were used. These V i v i d Autobiographical Memories 91 versions d i f f e r e d from one another only i n the order the questions were presented. Of p a r t i c u l a r i n t e r e s t here, was the placement of the vividness scale. In two versions, the vividness scale was placed at the beginning of the t e s t booklet; i n two i t was placed approximately i n the middle; i n the remaining two versions i t was placed toward the end. Space was also provided for the subject to indicate t h e i r present age, the age of the memory, and t h e i r sex. Although there was a great deal of v a r i a t i o n i n the time required f o r subjects to complete the enti r e experiment, no subject required longer than one h a l f hour. Results The memories c o l l e c t e d i n t h i s study ranged i n vividness ratings from 2 to 9. On comparing the s i x versions of the te s t booklet, no s i g n i f i c a n t order e f f e c t s were found f o r vividness ratings. The mean vividness r a t i n g was 7.32. Forty nine percent of the memories received vividness ratings of 7 or l e s s . Therefore, 7 provided the cut-off value f o r an approximate median s p l i t . Memories c l a s s i f i e d as VMs were defined as having vividness ratings of 8 or 9 (mean = 8.4, n = 104). Non-VMs were defined as having vividness ratings of less than 8 (mean = 6.0, n = 99). T-tests and c o r r e l a t i o n a l analyses were conducted between VMs and non-VMs for a l l v a r i a b l e s . The r e s u l t s appear i n Appendix D. T-test and c o r r e l a t i o n a l r e s u l t s for the variables of primary i n t e r e s t i n V i v i d Autobiographical Memories 92 t h i s study (vividness, rehearsal, PLT, ART, CTHEN, CNOW, and a l l d i s t i n c t i v e n e s s variables) are presented i n Tables 5 and 6. Distinctiveness and vividness. The most important fi n d i n g i n Experiment 3 was that some types of d i s t i n c t i v e n e s s were s i g n i f i c a n t l y correlated with vividness, whereas others were not. No s i g n i f i c a n t correlations were found between dSELF, dOTHER, dLOC, or dPPL, and vividness ratings. D i s t i n c t i v e emotions (dEM) as well as d i s t i n c t i v e thoughts (dTHT) and d i s t i n c t i v e actions (dACT) were, however, found to cor r e l a t e s i g n i f i c a n t l y with vividness (see Table 6). I t i s i n t e r e s t i n g to note that two of these variables are r e l a t i v e l y i n t e r n a l (cognitive-affective) events (dEM and dTHT). Therefore, two new composite variables were calculated. The f i r s t , COG-AFF, was calculated by adding together dTHT and dEM ratings. The s e c o n d — d i s t i n c t i v e external aspects—was calculated by adding the dLOC, dACT, and dPPL ratings together. Cognitive-Affective d i s t i n c t i v e n e s s was found to cor r e l a t e .36 with vividness; whereas, the composite r a t i n g of d i s t i n c t i v e external aspects correlated with vividness only .12. Therefore, although vividness i s associated with events that are distinguishable from other events by external c h a r a c t e r i s t i c s , d i s t i n c t i v e n e s s of cognitive and a f f e c t i v e c h a r a c t e r i s t i c s are better predictors of vividness. The percentages and mean vividness ratings of memory events c l a s s i f i e d as \" f i r s t \" , \" l a s t \" , \"only\", and \"one of a V i v i d Autobiographical Memories 93 s e r i e s \" were also calculated. The largest category was f i r s t -time events (40%), followed by only events (32%). One of a series events accounted for only 25% of the memories, while a mere four percent were c l a s s i f i e d as l a s t events. An analysis of variance of vividness ratings between a l l four groups revealed no s i g n i f i c a n t differences. Thus, i t appears that novel ( f i r s t and only) events, or perhaps temporally d i s t i n c t i v e events, are more l i k e l y to be reported, but they are not more l i k e l y to be v i v i d . The low frequency of l a s t -time events being reported i n t h i s study i s l i k e l y due to the two or more year memory age r e s t r i c t i o n . Excluding memories for events that took place i n the two years immediately preceding the study i s l i k e l y to eliminate more last-time events than f i r s t , only, or series events from the pool of memories the subjects could report. Vividness and ordinary memory theories. The Pearson c o r r e l a t i o n between covert rehearsal and vividness found i n Experiment 3 was .15. While t h i s c o r r e l a t i o n i s s i g n i f i c a n t (p_ = .015) i t i s much lower than the .40 c o r r e l a t i o n found between these same variables i n Experiment 2. In order to determine why t h i s might be so, scatterplots of rehearsal ratings from Experiments 2 and 3 were compared (see Figure 3a and 3b). The range of rehearsal ratings was r e s t r i c t e d i n Experiment 3 compared to Experiment 2. Only seven percent of the memories reported i n Experiment 3 were rated above 6 on the nine-point scale. While i t i s unclear why t h i s V i v i d Autobiographical Memories 94 r e s t r i c t i o n i n range occurred i n Experiment 3, i t i s c l e a r that the c o r r e l a t i o n found between vividness and rehearsal i s underestimated. Arousal at the time the event occurred (ART) was found to be s i g n i f i c a n t l y correlated with vividness, as were the following d i s t i n c t i v e n e s s v a r i a b l e s : dTHT, dEM, dACT, COG-AFF, and External (see Table 6). A stepwise multiple regression was conducted with rehearsal (RO), arousal then (ART), and a l l of the di s t i n c t i v e n e s s variables used as independent v a r i a b l e s . A multiple R of .38 was obtained. Only dTHT, dACT and dEM were found to be s i g n i f i c a n t independent predictors of vividness. Further, these factors were only able to predict 12% of the variance of the vividness ratings. Although the factors normally associated with superior memory performance that were included i n t h i s study (arousal, rehearsal and distinctiveness) can only account f o r twelve percent of the variance found i n vividness ratings, i t should be pointed out that a l l the events reported i n t h i s study were highly \"memorable\". The mean ra t i n g of subjects' confidence i n the accuracy of t h e i r memories was 7.6 on a nine-point scale. Subjects rated t h e i r confidence to be s i g n i f i c a n t l y higher f o r VMs (mean = 8.1) than for Non-VMs (mean = 7.2) and a s i g n i f i c a n t c o r r e l a t i o n was found between vividness and accuracy confidence (r = .42, p_ < .0001). I f confidence i n the accuracy of the memory can be used as a gauge of the V i v i d Autobiographical Memories 95 memorability of the event, then v i v i d memories can c e r t a i n l y be said to be memorable. Therefore, the memories e l i c i t e d i n t h i s study were a l l highly memorable r e s u l t i n g i n a r e s t r i c t e d range of memories being sampled. Therefore, the co r r e l a t i o n s between ordinary memory factors and vividness i s l i k e l y underestimated i n Experiment 3. In other words, even i f vividness i s highly associated with the same factors which underlie superior memory qu a l i t y i n general, t h i s r e l a t i o n s h i p would be underestimated i n t h i s study due to the f a c t that only well remembered events were included. Consequentiality. S i g n i f i c a n t c o r r e l a t i o n s were found between a l l of the consequentiality measures and vividness. The c o r r e l a t i o n among a l l possible p a i r s of \"consequentiality then\" v a r i a b l e s (EE, EC and EI) was .73. The cor r e l a t i o n s among consequentiality now (AE, AC,and Al) variables, ranged from .70 to .78. Although they were not e n t i r e l y redundant, i t seems these scales a l l indexed a s i m i l a r concept. Therefore, based on t h i s finding, we may conclude that expected enduring e f f e c t (EE) and actual e f f e c t (AE) scales used i n Experiments 1 and 2 assessed the same concept that previous researchers defined as consequentiality and importance. For ease of presentation, i t i s preferable to have only one variable defined as consequentiality. Therefore, two composite variables were calculated by adding together a l l three consequentiality \"then\" variables (CTHEN) and by adding together a l l three consequentiality \"now\" V i v i d Autobiographical Memories 96 variables (CNOW). As found i n Experiment 2, actual consequentiality (CNOW) was s l i g h t l y more p r e d i c t i v e of v i v i d r e c a l l than consequentiality at the time of occurrence (CTHEN) (see Table 6). The composite variables (CNOW and CTHEN) were found to pre d i c t vividness s l i g h t l y better than any of the in d i v i d u a l consequentiality variables alone. As i n Experiment 2, there was a large range of consequentiality ratings (3 to 26) found f o r v i v i d memories. Further, inspection of scat t e r p l o t s revealed that the re l a t i o n s h i p between CNOW and vividness and between CTHEN and vividness was continuous. Therefore, Experiment 3 re p l i c a t e s the finding of Experiment 2, by showing that although consequentiality i s rela t e d to vividness of r e c a l l , the rel a t i o n s h i p i s neither of the necessary nor of the continuous nature that i s implied by Brown and Kulik's \"Now P r i n t \" memory model. Unlike i n Experiment 2, p a r t i a l l i n g out the e f f e c t s of covert rehearsal (RC) did not reduce the c o r r e l a t i o n between vividness and consequentiality to a non-significant l e v e l . This f a i l u r e to re p l i c a t e i s l i k e l y due to the r e s t r i c t e d range of rehearsal ratings found i n Experiment 3. I t i s i n t e r e s t i n g to note though, that even with a r e s t r i c t e d range, rehearsal (RC) and CNOW correlated .51. Pleasantness. Unlike Experiment 2, i n which no s i g n i f i c a n t c o r r e l a t i o n was found between pleasantness ratings (PLT and PLN) and vividness, a s i g n i f i c a n t negative c o r r e l a t i o n was found between PLT and vividness i n Experiment V i v i d Autobiographical Memories 97 3 (see Table 6). This i s consistent with the over-reporting of unpleasant memories found i n Experiment 1. In order to see i f t h i s tendency f o r VMs to be more unpleasant was explainable i n terms of factors normally thought to a f f e c t memory quality , p a r t i a l c o r r e l a t i o n s were conducted c o n t r o l l i n g f o r rehearsal, arousal (ART), and the c o g n i t i v e - a f f e c t i v e d i s t i n c t i v e n e s s (AFF-COG) va r i a b l e s . When arousal or rehearsal were con t r o l l e d for, a s i g n i f i c a n t c o r r e l a t i o n remained between PLT and vividness. However, when Cog-Aff was co n t r o l l e d for, the c o r r e l a t i o n between vividness and PLT was reduced to a non-s i g n i f i c a n t l e v e l ( p a r t i a l r = .09, p_ > .05). Conversely, i f PLT i s co n t r o l l e d for, the c o r r e l a t i o n between vividness and COG-AFF remained s i g n i f i c a n t ( p a r t i a l r = .36, p_ < .001). A possible explanation of both the over-reporting of unpleasant memories found i n Experiment 1, and the c o r r e l a t i o n between pleasantness and vividness found i n Experiment 3, i s that unpleasant events are more l i k e l y to involve d i s t i n c t i v e thoughts and emotions, and, therefore, are more l i k e l y to be v i v i d l y r e c a l l e d . Sex Differences. Sex differences i n vividness ratings were assessed for a l l variables i n the study. The 69 males included i n the study were found to rate t h e i r memories as s i g n i f i c a n t l y less v i v i d than the 134 females (p_ <.05). The only other variables on which sex differences were found were arousal and d i s t i n c t i v e thoughts. Females rated t h e i r reported events to be s i g n i f i c a n t l y more arousing (p_ = .001) V i v i d Autobiographical Memories 98 and c o g n i t i v e l y d i s t i n c t i v e (p_ = .008). I t i s also i n t e r e s t i n g to note that when separate multiple regressions were preformed on the data for the women and men, d i f f e r e n t factors are found to predict vividness for each sex. For males, the only factor that s i g n i f i c a n t l y predicated vividness was arousal then (ART). This factor was found to account f o r approximately 9% of the variance. For females, three factors, d i s t i n c t i v e emotions (dEM), d i s t i n c t i v e a c t i v i t i e s (dACT), and d i s t i n c t i v e thoughts (dTHT) were found to be s i g n i f i c a n t independent predictors of vividness. In t o t a l these three factors accounted for approximately 21% of the variance i n the vividness ratings. A possible explanation for t h i s difference i s that men have a tendency not to report memories that are associated with events l i k e l y to have the d i s t i n c t i v e factors which underlie v i v i d r e c a l l . Evidence consistent with t h i s p o s s i b i l i t y comes from a study by De Vries (1988) i n which subjects were asked to report memories for important events i n t h e i r l i v e s . De Vries found that \"women i d e n t i f i e d a greater number of events and discussed them more i n terms of connectedness with s i g n i f i c a n t others than did men\" (p. i i i ) . Had content been controlled for t h i s sex difference found i n Experiment 3 may well have been eliminated. Persistence and Vividness. Memory age ranged from 2 to 40 years. The mean memory age was 5.8 (mode = 3.0, median = 4.0). Memory age and vividness were not s i g n i f i c a n t l y correlated. As i n Experiment 2, there was no s i g n i f i c a n t V i v i d Autobiographical Memories 99 difference i n memory age between VMs and non-VMs. The difference i n memory age-range was thought to be due to methodological differences i n e l i c i t i n g the memory protocols. In Experiment 2 , when subjects were asked to report any memory that occurred two or more years ago, the memory age requirement appears to have biased subjects toward reporting a memory that occurred j u s t about two years previously. In Experiment 3 , subjects were t o l d on the Information/Consent sheet, that they would be asked to report a personal memory. They were not informed of the memory age requirement u n t i l they read the actual t e s t i n s t r u c t i o n s . Therefore, the fac t that a larger memory age-range was found i n Experiment 3 than i n Experiment 2 , may r e f l e c t the fact that subjects i n Experiment 3 had already selected a target memory p r i o r to reading about the memory age requirement. In other words, Experiment 3 subjects may have been less l i k e l y to search for a reportable memory by asking \"What was I doing two years ago?\". In contrast, t h i s may have been a major strategy for subjects i n Experiment 2 . The difference i n r e s u l t s i l l u s t r a t e s the importance of precise cueing procedures i n autobiographical memory studies i n general. Discussion The main finding from Experiment 3 , that events that involve d i s t i n c t i v e cognitive and a f f e c t i v e aspects are more l i k e l y to produce VMs, supports the idea that ordinary memory V i v i d Autobiographical Memories 100 processes can be used to explain v i v i d r e c a l l . I f the c o g n i t i v e - a f f e c t i v e components of the event are distinguishable and are, therefore, e a s i l y retrieved, they can, once r e c a l l e d , serve as cues f o r other aspects of the event. Remembering one's feelings and thoughts may be a l l that i s required i n order that the memory representation be r e c a l l e d as though one were \" r e l i v i n g \" the event. This, i n fa c t , may explain an observation made by Neisser (1982), when he noted that: At f i r s t glance, flashbulb memories appear to be obvious i l l u s t r a t i o n s of a commonsense p r i n c i p l e : people remember what i s important. The problem a r i s e s only when we r e a l i z e (with Brown and Kulik) that the actual content of flashbulb memories does not meet that c r i t e r i o n . The fact of John F. Kennedy 1s death may be important, but what difference does i t make who t o l d me about i t , or where I was when I heard the news? None, r e a l l y . Why, then, do I seem to remember such things? (p. 43) The answer to t h i s question could be that the most d i s t i n c t i v e and s a l i e n t aspects of the event are the emotions and thoughts experienced. These aspects, once r e c a l l e d , cue memories for ad d i t i o n a l aspects of the event. Because the cues used to e l i c i t the d e t a i l s of the event are personal and i n t e r n a l , the memory should be r e c a l l e d from t h i s same personal perspective. In other words, i t i s the experiencing of the event, not the facts of the event, which i s r e c a l l e d . The second finding of Experiment 3, that only about 12% of the variance found i n the vividness ratings was accounted for by ordinary memory theory variables (rehearsal, arousal V i v i d Autobiographical Memories 101 then, and various measures of d i s t i n c t i v e n e s s ) , casts some doubt upon the a b i l i t y of ordinary memory theories to explain VMs. Three indicators suggest that the magnitude of t h i s r e l a t i o n s h i p i s underestimated. F i r s t , s c a t t e r p l o t analysis revealed that there was a r e s t r i c t i o n i n the range of rehearsal r a t i n g i n Experiment 3. In Experiment 2, where such a r e s t r i c t i o n did not occur, the c o r r e l a t i o n between vividness and rehearsal was much higher. I t i s unclear why t h i s r e s t r i c t i o n occurred i n Experiment 3. Second, there was a general r e s t r i c t i o n i n the memorability- range of the events reported. Because subjects were allowed to choose any memory to report, i t i s l i k e l y that they selected events that were well remembered. I f the sample included memories for events which the subjects did not remember very well, the re l a t i o n s h i p between vividness and factors which are generally associated with superior memory performance would l i k e l y be higher. Third, extraneous variables may have systematically affected the content of the memories subjects chose to report (ie, sex differences, pleasant vs unpleasant memories). Co n t r o l l i n g f o r memory content, therefore, may eliminate the influence of various extraneous variables. The t h i r d f i nding i n Experiment 3, that the r e l a t i o n s h i p found between vividness and the e f f e c t variables (EE and AE), generalized to d e f i n i t i o n s of consequentiality used by other researchers, provides further empirical evidence that the c h a r a c t e r i s t i c s of v i v i d l y r e c a l l e d , persistent memories are V i v i d Autobiographical Memories 102 not of the nature predicated by Brown and Kulik's \"Now P r i n t \" s p e c i a l mechanism theory. Unlike the low c o r r e l a t i o n between importance and consequentiality ratings found by Rubin and Kozin, the r e s u l t s of Experiment 3 found that a l l three d e f i n i t i o n s of consequentiality used i n t h i s study were highly correlated with each other. Therefore, the f a i l u r e i n Experiments 2 and 3 to f i n d a discontinuous, necessary r e l a t i o n s h i p between expected e f f e c t and vividness i s not due to EE indexing some concept other than that which Brown and Kulik and other researchers (Rubin & Kozin, 1984; Cohen & Faulkner, 1984; Conway & Berkerian, 1988) defined as consequentiality. The nature of the rela t i o n s h i p s among a l l consequentiality variables i n Experiment 3 r e p l i c a t e d those found i n Experiment 2. Consequentiality, therefore, does not appear to be rel a t e d to vividness i n the discontinuous, necessary manner required by Brown and Kulik's theory. General Discussion Overall, the three experiments presented i n t h i s thesis addressed four main questions. F i r s t , are some l i f e - e v e n t s are more l i k e l y to be v i v i d l y r e c a l l e d than others? Second, can the \"Now P r i n t \" s p e c i a l memory mechanism theory proposed by Brown and Kulik explain the existence of VMs? Third, can ordinary memory theories explain the existence of VMs? Fourth what aspects of di s t i n c t i v e n e s s allow the memory system to V i v i d Autobiographical Memories 103 d i f f e r e n t i a t e between r e c a l l of events i n a v i v i d versus non-v i v i d manner? The f i r s t question, whether or not some events are more l i k e l y to be v i v i d l y r e c a l l e d , was explored by categorizing reported memories into thematic c l u s t e r s . Thematic categories were derived from subjects 1 reported VMs i n Experiment 1. The percentage of reported memories f a l l i n g into each category was compared to the average percentage of r e a d i l y a v a i l a b l e memories subjects estimated they had for each category. The percentage of VMs reported for accidents, i n j u r i e s , i l l n e s s e s , and operations was found to be s i g n i f i c a n t l y higher than the average percentage of r e a d i l y available memories f o r accidents, i n j u r i e s , i l l n e s s e s , and operations. The percentage of VMs reported for vacation memories, on the other hand, was found to be s i g n i f i c a n t l y lower than the percentage of r e a d i l y a v a i l a b l e vacation memories subjects estimated they possessed. Further, the number of VMs reported for a l l other thematic categories were found to be proportional to the average estimated number of re a d i l y available memories for those categories (see Figure 2). This finding suggests some factor(s) generally associated with injurious events, but not with vacation events, underlies v i v i d r e c a l l . I t was unclear, however, whether t h i s finding r e f l e c t e d a true difference i n the thematic content of VMs and non-VMs, or was an a r t i f a c t of the d i f f e r e n t methods used to assess thematic category percentages for VMs and re a d i l y available memories. In other V i v i d Autobiographical Memories 104 words, i t could be that, when asked to write a v i v i d memory account, subjects tended more to report memories f o r injurious events and tended less to report memories of vacations even though the number of VMs they could have reported were proportional to the number of r e a d i l y a v a i l a b l e memories i n each category. In order to eliminate t h i s possible explanation, t h i s same issue was explored i n Experiment 2. In Experiment 2, subjects were asked to report a memory and then to rate i t for vividness. The percentages of reported VMs were compared to percentages of reported non-VMs i n each thematic category. The r e s u l t s of t h i s comparison showed that both l i f e - t h r e a t e n i n g and injurious events were more l i k e l y to produce VMs than non-VMs. The reverse was found f o r vacation memories. The numbers of VMs reported f o r a l l other thematic categories were found to be proportional to the numbers of reported non-VMs for each category (see Figure 3). This f i n d i n g r e p l i c a t e s the findings of Experiment 1 and suggests that some factor(s) associated with l i f e - t h r e a t e n i n g and i n j u r i o u s events, but not with vacations, can systematically a f f e c t the l i k e l i h o o d of an event being v i v i d l y r e c a l l e d . The second question addressed was whether or not Brown and Kulik's \"Now P r i n t \" special memory mechanism theory can account f o r the existence of VMs. The finding that l i f e -threatening and injurious events were more l i k e l y to be v i v i d l y r e c a l l e d i s consistent with the suggestion made by V i v i d Autobiographical Memories 105 Pettigrew (1978) that a special mechanism that accurately and permanently encodes memories for l i f e - t h r e a t e n i n g events may have evolved because i t has s u r v i v a l value. To examine t h i s issue, the re l a t i o n s h i p s between vividness and the two primary factors (consequentiality and a f f e c t i v e arousal) that Brown and Kulik proposed as t r i g g e r s for the spe c i a l mechanism, were examined. Consequentiality was defined i n Experiment 2 as expected and actual e f f e c t . Although both of these factors were s i g n i f i c a n t l y correlated with vividness i n Experiment 2, the nature of the r e l a t i o n s h i p between these factors and vividness was not found to be of the discontinuous and necessary nature required by the \"Now P r i n t \" model. In order to eliminate the p o s s i b i l i t y that expected and enduring e f f e c t were not indexing the same concept that Brown and Kulik refer r e d to as consequentiality. Subjects were required, i n Experiment 3, to rate t h e i r memories on scales for actual and expected consequentiality and importance as well as for actual and expected e f f e c t . The relationships among a l l three scales and vividness were found to r e p l i c a t e both the degree and nature of the r e l a t i o n s h i p found between the e f f e c t variables and vividness i n Experiment 2. Brown and Kulik's model also s t i p u l a t e d that only factors that are present at the time of encoding could t r i g g e r the spec i a l mechanism. In both Experiments 2 and 3, vividness was found to be equally correlated with expected and actual consequentiality. Therefore, factors that are not present at V i v i d Autobiographical Memories 106 the time the event occurred, are correlated with the vividness with which the event i s r e c a l l e d . Evidence that further supports the idea that Brown and Kulik's model can not explain the existence of VMs, i s the find i n g i n Experiment 2, that the re l a t i o n s h i p between consequentiality and vividness could be explained i n terms of rehearsal. I t should be noted, however, that the ratings used i n a l l three studies were retrospective. I f a repeated measures design were used to assess expected consequentiality at the time the event a c t u a l l y occurred, d i f f e r e n t r e s u l t s might be found. The t h i r d question addressed i n t h i s t h e s i s i s whether or not ordinary memory theories can explain the existence of VMs. In order to address t h i s question, subjects were asked, i n both Experiments 2 and 3, to rate t h e i r memories on scales which indexed factors generally thought to underlie superior memory performance (rehearsal and arousal). Each of these factors was found to be s i g n i f i c a n t l y correlated with vividness; however, a great deal of the variance i n vividness ratings could not be explained by these factors. I t should be noted that, based on the confidence subjects reported having i n the accuracy of t h e i r memory accounts, very few, i f any, of the events reported were not well-remembered. The influence of those factors generally associated with superior memory performance was, therefore, assessed only for well-remembered events. Because i t i s reasonable to assume that VMs are well-V i v i d Autobiographical Memories 107 remembered, r e s t r i c t i n g the memorability-range would reduce the r e l a t i o n s h i p found between ordinary memory factors and vividness. Using a procedure that e l i c i t s memories over a large memorability-range, would more c l e a r l y assess the re l a t i o n s h i p between factors that underlie superior memory i n general and vividness. Also, because one of the main purposes of t h i s study was to examine the thematic content of the memories, content was not controlled f o r . As can be seen from the sex difference analysis conducted i n Experiment 3, the contents of memories may greatly influence the factors that are found to be most c l o s e l y associated with vividness. I t i s only reasonable to assume that VMs, i f not produced by a s p e c i a l mechanism, would be associated with the factors that influence memory performance i n general. In order for VMs to be explained within ordinary memory theories, however, i t must also be determined what factors d i f f e r e n t i a t e w e l l -remembered VMs from well-remembered non-Vms. The fourth issue addressed i n t h i s thesis, i s whether d i s t i n c t i v e n e s s of ce r t a i n aspects of the event, s p e c i f i c a l l y cognitive and a f f e c t i v e aspects, serve as d i f f e r e n t i a t i n g factors. Distinctiveness of a f f e c t i v e and cognitive aspects of the event were found to be s i g n i f i c a n t l y associated with vividness of r e c a l l , whereas, other measures of di s t i n c t i v e n e s s were not. The a b i l i t y to d i s t i n g u i s h an event from other s i m i l a r events that have occurred i n one's l i f e , may be associated with superior memory performance i n general. However, V i v i d Autobiographical Memories 108 d i s t i n c t i v e n e s s of thoughts and emotions appear to produce the cl e a r , l i f e - l i k e r e c a l l that d i f f e r e n t i a t e s VMs from w e l l -remembered non-VMs. This finding i s also consistent with the thematic content findings. Whereas looking at the Grand Canyon may be a d i s t i n c t i v e and memorable event i n one's l i f e , i t i s not generally associated with the a f f e c t i v e and c o g n i t i v e l y d i s t i n c t experience of thinking \"*@#&, I'm gonna to d i e ! ! \" V i v i d Autobiographical Memories 109 References B a r t l e t t , F. C. (1932). Remembering. New York: Cambridge Unive r s i t y Press. Brewer, W. F. (1988). Memory for randomly sampled autobiographical events. In U. Neisser and E. Winograd (Eds.), Remembering reconsidered: Ecological and t r a d i t i o n a l approaches to the study of memory. (pp. 21-90). New York: Cambridge University Press. Brewer, W. F. (1986). What i s autobiographical memory? In D.C. Rubin (ed), Autobiographical memory, (pp. 22-49). New York: Cambridge University Press. Brown, R. & Kulik, J . (1977). Flashbulb memories. Cognition. 5, 73-99. Cohen, G. & Faulkner, D. (1988). L i f e span changes i n autobiographical memory. In M. M. Gruneberg, P.E. Morris & R. N. Sykes (eds.), P r a c t i c a l aspects of memory: Current research and issues. Vol. 1. Memory i n everyday l i f e , (pp.277-282). Chichester: John Wiley and Sons. Colgrove, F. W. (1899). Individual memories. American Journal of Psychology. 10, 228-255. Conway, M. A. & Berkerian D. A. (1988). C h a r a c t e r i s t i c s of v i v i d memories. In M. M. Gruneberg, P.E. Morris & R. N. Sykes (eds.), P r a c t i c a l aspects of memory: Current research and issues. Vol. 1. Memory i n everyday l i f e , (pp. 519-524). Chichester: John Wiley and Sons. De Vries, B. (1988). The concept of s e l f i n a l i f e - s p a n l i f e event contexts Unpublished doctoral d i s s e r t a t i o n , University of B r i t i s h Columbia, Vancouver. Ebbinghaus, H. (1964). Memory. New York: Dover. (Original work published 1885) Larsen, S. F. (1988). Remembering reported events: Memory for news i n ecolo g i c a l perspective. In M. M. Gruneberg, P.E. Morris & R. N. Sykes (eds.), P r a c t i c a l aspects of memory: Current research and issues. Vol. 1. Memory i n everyday l i f e , (pp. 440-445). Chichester: John Wiley and Sons. Linton, M. (1975). Memory for real-world events. In D. A. Norman & D. E. Rumelhart (Eds.), Explorations i n Cognition (pp. 376-404). San Francisco: Freeman. V i v i d Autobiographical Memories 110 Livingston, R. B. (1967), Reinforcement. In G. C. Quarton, T. Melnechuck, &. F. O. Schmitt (eds.), The neurosciences; A study program, (pp. 568-576). New York: Rockefeller Uni v e r s i t y Press. Mehrabian, A. & Russell, J . A. (1974). An Approach to Environmental Psychology. Cambridge, MA: MIT Press. McCloskey, M., Wible, C. G. & Cohen, N. J . (1988). Is there a sp e c i a l flashbulb-memory mechanism? Journal of Experimental Psychology: General. 117(2), 171-181. Neisser, U. (1982). Snapshots or benchmarks? In U. Neisser (Ed.), Memory observed: Remembering i n natural contexts (pp. 43-48). San Francisco: Freeman. Neisser, U. (1986). Remembering Pearl Harbor: Reply to Thompson & Cowan. Cognition. 23. 285-286. Pettigrew, J . D. (1978). The locus coeruleus and c o r t i c a l p l a s t i c i t y . Trends i n Neurosciences. 1, 73-74. Pillemer, D. B. (1984). Flashbulb memories of the assassination attempt on President Reagan. Cognition. 16, 63-80. Rabbitt, P. & Winthorpe, C. (1988). What do old people remember? The Galton paradigm reconsidered. In M. M. Gruneberg, P.E. Morris & R. N. Sykes (eds.), P r a c t i c a l aspects of memory: Current research and issues. Vol. 1. Memory i n everyday l i f e , (pp. 303-309). Chichester: John Wiley and Sons. Rubin, D.C, & Kozin, M. (1984). V i v i d Memories, Cognition. 16, 81-95. Sheingold, K. & Tenney, Y.J. (1982). Memory for a s a l i e n t childhood event. In U. Neisser (Ed.), Memory observed: Remembering i n natural contexts, (pp. 201-212). San Francisco: Freeman. Strube, G. & Neubauer, S. (1988). Remember that exam? In M. M. Gruneberg, P.E. Morris & R. N. Sykes (eds.), P r a c t i c a l aspects of memory: Current research and issues. Vol. 1. Memory i n everyday l i f e , (pp. 247-252). Chichester: John Wiley and Sons. Thompson, C. P. & Cowan, T. (1986). Flashbulb memories: A nicer i n t e r p r e t a t i o n of a Neisser r e c o l l e c t i o n . Cognition. 22., 199-200. V i v i d Autobiographical Memories 111 von Restorff, H. (1933). Uber die wirking von bereichsbildungen im Spurenfeld. In W. Kohler & H. von Restorff, Analyse von Vorgangen im Spurenfeld. I. Psycholocfische Forschung, 18, 299-342. Winograd, E. (1988). Continuities between Ecolo g i c a l and Laboratory Approaches to Memory. In U. Neisser and E. Winograd (Eds.), Remembering reconsidered: E c o l o g i c a l and t r a d i t i o n a l approaches to the study of memory. (pp. 11-20). New York: Cambridge University Press. Winograd, E., & K i l l i n g e r W. A. J r . (1983). Relating age at encoding of early childhood to adult r e c a l l : Development of flashbulb memories. Journal of Experimental Psychology: General. 112.(3), 413-422. Yarmey, A. D. & B u l l , M. P. I I I . (1978). Where were you when President Kennedy was assassinated? B u l l e t i n of the Psychonomic Society. 11.(3), 133-135. Y u i l l e , J.C. & Cutshall, J.L. (1986). A case of eyewitness memory of a crime, Journal of Applied Psychology. 71. 291-301. V i v i d Autobiographical Memories 112 Table 1 Experiment 1. T--tests between pleasant and unpleasant event memories. Pleasant Unpleasant P means means Subject age 20.8 20.3 ns Memory age 6.4 8.1 <.01 EE 4.9 4.9 ns AE 4.4 3.8 <.01 Rehearsal 3.7 3.3 <.02 EE = expected enduring e f f e c t AE = actual e f f e c t V i v i d Autobiographical Memories 113 Table 2 Experiment 2. T-Tests between v i v i d and non-- v i v i d memories. V i v i d Non-vivid P means means n = 119 n = 140 Rehearsal 3.05 2.35 <.001 EE 4.16 3.62 .025 AE 3.98 3.24 .001 PLT -0.57 2.27 ns PLN 4.34 5.49 ns ART 13.89 11.54 .015 ARN 3.91 1.41 .012 INTT 5.95 5.45 <.001 INTN 4.33 3.54 <.001 Absolute PLT 13.60 12.32 .ns Absolute PLN 13.25 11.04 .003 EE = expected enduring e f f e c t AE = actual e f f e c t PLT = pleasantness then PLN = pleasantness now ART = arousal then ARN = arousal now INTT = i n t e n s i t y then INTN = i n t e n s i t y now T a b l e 3 E x p e r i m e n t 2. P e a r s o n C o r r e l a t i o n s . V IV ID REH. EE AE PLT PLN ART ARN INTT INTN absPLT absPLN V I V I D 1.00 REHEARSAL .40 1.00 EE .24 .39 1.00 AE .26 .47 .46 1.00 PLT -.04 .07 .05 .18 1.00 PLN -.01 .02 .07 .07 .74 1.00 ART .18 .01 .12 -.01 .02 .06 1.00 ARN .20 .23 .13 .23 .11 -.01 .30 1.00 INTT .24 .17 .38 .16 -.17 -.07 .42 .29 1.00 INTN .33 .44 .26 .46 .16 -.02 .14 .44 .34 1.00 absPLT .10 .10 .20 .20 .15 .10 .15 .14 .20 .11 1.00 absPLN . 22 .30 .18 .34 .41 .50 .11 .21 .12 .28 .43 1.00 n = 259 c r i t i c a l r p. <.05 = .12 p. <.01 = .16 p. < .001 = .21 EE = e x p e c t e d e n d u r i n g e f f e c t AE = a c t u a l e f f e c t PLT = p l e a s a n t n e s s t h e n PLN = p l e a s a n t n e s s now ART = A r o u s a l t h e n ARN = a r o u s a l now INTT = i n t e n s i t y t h e n INTN = i n t e n s t i t y now V i v i d Autobiographical Memories 115 Table 4 Experiment 2. P a r t i a l c o r r e l a t i o n c o e f f i c i e n t s with vividness ( c o n t r o l l i n g f or rehearsal) REHEARSAL EE r .10 p .12 AE .09 . 16 PLT -.07 .23 PLN -.02 .74 ART .20 .002 ARN .12 .06 INTT .20 .002 INTN .18 .002 (c o n t r o l l i n g f or EE and AE) REHEARSAL EE AE PLT PLN ART ARN INTT INTN r .31 - -.09 -.04 . 18 . 14 . 18 .24 p <.001 .09 .26 .002 .01 .002 <.001 EE = expected enduring e f f e c t AE = actual e f f e c t PLT = pleasantness then PLN = pleasantness now ART = arousal then ARN = arousal now INTT = i n t e n s i t y then INTN = i n t e n s i t y now V i v i d Autobiographical Memories 116 Table 5. Experiment 3. T-test analyses. V i v i d Non-vivid p means* means* n = 104 n = 99 MEMORY AGE 5.5 6.2 ns PLEASANT THEN -1.1 2.4 ns AROUSAL THEN 14.9 13.4 ns CTHEN 14.6 12.1 <.01 CNOW 15.1 11.9 <.001 REHEARSAL COVERT 4.0 3.4 <.05 DISTINCT TO SUBJECT 6.0 5.6 ns DISTINCT FROM OTHERS 5.7 4.9 ns DISTINCTIVE THOUGHTS 6.5 5.7 <.05 DISTINCTIVE EMOTIONS 7.8 6.9 <.001 DISTINCTIVE PEOPLE 7.8 7.4 ns DISTINCTIVE LOCATION 7.6 7.3 ns DISTINCTIVE ACTIVITIES 7.5 7.4 ns CTHEN = consequentiality then CNOW = consequentiality now * note. Higher ratings indicate greater l e v e l s of the var i a b l e or stronger agreement with the va r i a b l e statement. Table 6. VIVID PLT ART CTHEN CNOW RC dTHT dEM dLOC dPPL dACT COG-AFF EXT VIVID 1.00 PLEASANT THEN -.12 1.00 AROUSAL THEN .20 .03 1.00 CTHEN .23 .11 .27 1.00 CNOW .26 .05 .11 .50 1.00 REH. COVERT .15 .00 .07 .51 .29 1.00 THOUGHTS .26 - .15 .17 .14 .23 .12 1.00 EMOTION .35 -.03 .30 .28 .34 .16 .35 1.00 LOCATION .01 .01 .02 .03 .10 .01 .00 - .02 1.00 PEOPLE .08 .11 .05 .02 .02 .03 .07 .02 .26 1.00 ACTIONS .16 .10 .28 .04 .01 .07 .18 .08 .22 .15 1. 00 COG-AFF .36 .12 .28 .35 .25 .13 .87 .77 .07 .02 04 1.00 EXTERNAL .12 .12 .17 -.06 .02 .02 .06 .04 . 68 .71 67 .06 1.00 < H-< U l . > c n-o cr H\" O vQ hi PJ *d o1 (n = 203) C r i t i c a l r. (p_ = .05 .12) Ip. =.01 .16) (p_ = .001 .21) one t a i l e d . O PLT = pleasantness then ART = arousal then £L CTHEN = composite consequentiality then scores CNOW = composite consequentiality now scores jjf RC = covert rehearsal dTHT = di s t i n c t i v e thoughts 3 dEM = d i s t i n c t i v e emotions dLOC = di s t i n c t i v e locations dppl = d i s t i n c t i v e people dACT = di s t i n c t i v e a c t i v i t i e s COG-AFF = composite dTHT and dEM scores EXT = composite dLOC, dPPL, and dACT scores H V i v i d Autobiographical Memories Figure 1 Experiement 1. Thematic Category Percentages W////////A « 2 « 2 vacations pi child LMs unpl adult LMs death/hearing of beg. relationship moving nature related embarrassment family problems car/no injuries end relationships witness accident mjjj^^ life threatening wj^jjjjj^^ births breaking rules physical fights pi adult LMs unpl. child LMs 12.5 10.4 accidents, injuries 10,6 -+ 8 10 12 percentage I VIVID BASEL INE 18 20 LMs = landmarks V i v i d Autobiographical Memories 119 Figure 2 Experiment 2. Thematic Category Percentages witnessing accident accident, injury etc. life-threatening vacations physical fights embassasment hearing bad news -^'f beg. relationship end relationship deaths breaking rules family problems unpleasant other pleasant other . unpleasant LMs fefl.Z pleasant LMs VIVID NON-VIVID 10 15 20 percentage * Catagories containing less than 2% total memories not presented. Figure 3a Experiment 2 V i v i d Autobiographical Memories 120 Scatterplot of Rehearsal with V i v i d 1 7 * 6 * .8% 5 * 1.5% 2% 3% 4 * 8% 7% 3% 3 .8% * .8% 9% 12% 2.5% 2 .8% 2% 3% 10% 11% 2% 1 1.5% 3.5% 5% 6.5% 4% 1 2 3 4 V i v i d 5 6 7 = 1 case = approx 0.4% n = 259. Figure 3b Experiment 3 V i v i d Autobiographical Memories 121 Scatterplot of Rehearsal with V i v i d e 9 8 7 * 1.5% * 6 1% 1% 2% * 5 * 1.5% 3% 4% 3.5% 4 2% 4.5% 2% 3 * * 1% 4.5% 4.5% 6% 2.5% 2 1% . * 1% 2.5% 8.5% 7.5% 4% 1 * 2% * 3% 8 . 5% 7% 6.5% 2 1 case 3 4 = approx 0.5% 5 V i v i d 6 n = 7 205 8 9 V i v i d Autobiographical Memories 12 2 Appendix A Memory Checklist Please Read Directions c a r e f u l l y before beginning!! This questionnaire i s designed to assess the types of events for which you have r e a d i l y available memories, (memories which come e a s i l y to mind). Please read through the following categories. I f you f i n d that you have a memory for an incident which f a l l s into a c e r t a i n category place a \"1\" (one) i n the space provided to the r i g h t of the relevant item. I f more than one incident which f a l l s into a category cames e a s i l y to mind then mark the number of such memories you have which are relevant. I f no memories come to mind e a s i l y , put a dash (-) i n the space to the r i g h t of the item. 1. L i f e threatening experiences involving: I) i l l n e s s II) drowning III f i r e IV) f a l l i n g V) i n f l i c t e d violence VI) other 2 . B i r t h of family member: I) s i b l i n g II o f f s p r i n g III) other family . 3. Being involved i n a physical f i g h t i n which you were: I) aggressor II) the v i c t i m III both to blame . 4 . Re-locating or moving: I) to new house (same ci t y ) II) new c i t y III) new province or state (same county) . IV) new country . 5. Vacations to: I) within B r i t i s h Columbia II) elsewhere i n Canada III) Washington state IV) Hawaii V) Disneyland VI) elsewhere i n USA VII) Mexico VII Europe VIII) other . 6. Being involved i n an incident i n which you were embarrassed: I) by yourself II) by a s i b l i n g III) by a parent IV) by a teacher V) other . V i v i d Autobiographical Memories 12 3 7. Getting caught breaking rules or laws: I) school related II) recreation related III) alcohol related IV) t h e f t V) tobacco r e l a t e d VI other 8. F i r s t hearing bad news: I) death of s i b l i n g II) death of parent III) death of grandparent IV) death of other family member V) death of f r i e n d VI death of pet VII other unpleasant events 9. Family problems: I) parental separation or divorce_ II) parental arguments III) arguments with parents IV) separation from family members V) unusual punishments by parents 10. End of r e l a t i o n s h i p s : I) marital II) romantic III) friendship 11. Car incidents without i n j u r i e s : I) learning to drive_ II taking drivers t e s t near accidents IV minor accidents V) other 12. Unpleasant childhood Landmarks: I) getting l o s t II) school problems III) unpleasant sexual encounters IV) causing injury to another by accident V) being threatened of or fearing strangers VI) discomfort of fear i n strange environments VII) other 13. Pleasant childhood landmarks: (age 13 and Below) I) mastering a new s k i l l II) r eceiving award for sports III) r e c eiving an academic award IV) receiving recognition for musical t a l e n t s V) s o c i a l occasions (birthdays, bar mitzvahs, baptisms, f i r s t day school VI other 14. Unpleasant adult landmarks: (Age 14+) I) school trouble II) job related III) unpleasant sexual encounters IV) causing injury to other by accident V) other_ 15. Pleasant adult landmarks: (Age 14+) I) high school graduation II) sports related III) p a r t i e s , s p e c i a l events IV) f i r s t car V) f i r s t day at new job VI) f i r s t day at u n i v e r s i t y VII) pleasant sexual encounters VIII) other 16. Witnessing an accident: (In which you were not involved.) I) involving a car II) sports related III) involving a b i c y c l e of t r i c y c l e IV) other V i v i d Autobiographical Memories 124 17. Nature related incidents: I) involving animals II) unpleasant whether conditions III) pleasant whether or scenery 18. Beginning of r e l a t i o n s h i p s : I) romantic II) friendship III) reunion of old friendship IV) other 19. Accident with i n j u r i e s , i l l n e s s e s , operations (which you personally experienced). I) car accidents_ II) sports III) b i c y c l e / t r i c y c l e IV) operations V) i l l n e s s e s VI) other V i v i d Autobiographical Memories 125 Appendix B Experiment 2. Questionnaire. In the space provided on t h i s page, please describe b r i e f l y a personal episode (something i n which you yourself were involved) that occurred two years ago or longer. The episode should be of a short duration covering no more than one day. Student number . Present age . The following questions r e f e r to the episode which you recorded on page 1. 1. How old were you when t h i s event took place? 2. When t h i s event occurred, how much of an enduring e f f e c t did you expect i t to have on your l i f e ? no e f f e c t medium e f f e c t great e f f e c t 1 2 3 4 5 6 7 3. Please use the following scales to describe your feelings at the time. Some of the word pairs may seem unusual, but y o u ' l l probably f e e l more one way than another. For each p a i r below, put a checkmark. (example : : :) When you have finish e d , please be sure that there i s one check on each l i n e . Unhappy : :_ : : : : : : Happy Relaxed : : : : : : : : Stimulated Pleased : : : : : : : :__ Annoyed Excited : : : : : : : : Calm Uns a t i s f i e d : : : : : : : : S a t i s f i e d Sluggish : : : : : : : : Frenzied Contented : : : : : :_ : : Melancholic J i t t e r y : : : : : : : : Dull Despairing : : : : : : : : Hopeful Sleepy : : : : : : : : Wide awake Relaxed : : : : : : : : Bored Aroused : : : : : : : : Unaroused 4. Some events involve mixed feelings. In the space provided please indicate at the time the event occurred what percentage of the emotion you were experiencing was; a) % unpleasant b) % pleasant V i v i d Autobiographical Memories 126 5. Regardless of the type of emotion involved, how intense would you rate the emotions you were experiencing at the time of the event? not very intense intense 6. How much of an enduring e f f e c t do you f e e l t h i s event has a c t u a l l y had on your l i f e ? no e f f e c t medium e f f e c t great e f f e c t 7. Please use the following scales i n the same manner as on page 2 to describe your feelings when you think about the event. When you have finish e d , please be sure that there i s one check on each l i n e . Unhappy : : : : : : : : Happy Relaxed : : : : _: : : : Stimulated Pleased : : : : : : : : Annoyed Excited : : : : __: : : : Calm Un s a t i s f i e d : : : : : : : : S a t i s f i e d Sluggish : : : : : : : : Frenzied Contented : : : : : : : : Melancholic J i t t e r y : _: : : : : : : Dull Despairing : : : :_ : _: : : Hopeful Sleepy : : : : _: : _: : Wide awake Relaxed : : : : : : : : Bored Aroused : : : : : : : : Unaroused 8. Some memories involve mixed feel i n g s . In the space provided please indicate what percentage of the emotions you experience when remembering t h i s event are; a) % unpleasant b) % pleasant 9. Regardless of the type of emotion involved, how intense would you rate the emotions you experience when r e c a l l i n g t h i s event? not very intense intense 1 2 3 4 5 6 7 V i v i d Autobiographical Memories 127 10. When you think about t h i s event how v i v i d i s the memory? very fuzzy almost l i k e not c l e a r r e l i v i n g the event 1 2 3 4 5 6 7 11. How often do you think about t h i s event? almost almost never d a i l y 1 2 3 4 5 6 7 Please read through the following categories and indicate (by placing an X i n the space provided) the one which the event you provided on page one best f a l l s into. Please choose only one. 1. Witnessing an accident 2. Accident, injury, i l l n e s s , or operation which you experienced. 3. L i f e threatening event (an event where you were a f r a i d the outcome would be your own death. 4. B i r t h of a family member. 5. Relocating (Moving to a new county, c i t y or new home.) 6. Vacation 7. Physical f i g h t 8. Embarrassing incident 9. Unpleasant landmark* 10. Pleasant landmark* 11. Pleasant experiences not associated with landmarks 12. Unpleasant experiences not associated with landmarks 13. F i r s t hearing bad news (other than deaths of s i g n i f i c a n t others) 14. Beginning of a re l a t i o n s h i p 15. End of a re l a t i o n s h i p 16. Death of a s i g n i f i c a n t other 17. Car accidents which did not r e s u l t i n i n j u r i e s to you. 18. Nature related events (either pleasant or unpleasant) 19. Being caught breaking the law or rules. 20. Family problems * A land mark i s a turning point or an event that marks the culmination of a series of events. Examples of these are birthdays, weddings, high school graduation or winning an important sporting event. V i v i d Autobiographical Memories 128 Appendix C Experiment 3 questionnaire The following questions r e f e r to the event which you recorded on the previous page. Please place a checkmark i n the space which best describes your experience. Vividness When you think about t h i s event how v i v i d i s the memory? very fuzzy l i k e r e l i v i n g not c l e a r : : : : : : : : the event Rehearsal How often do you discuss t h i s event? almost almost never : : : : : : : : d a i l y How often do you think about t h i s event? almost almost never : : : : : : : : d a i l y A f f e c t and Arousal Please use the following scales to describe your feelings at the time the event took place. Some of the word p a i r s may seem unusual, but y o u ' l l probably f e e l more one way than another. For each p a i r below, put a checkmark. When you have fi n i s h e d , please be sure that there i s one check on each l i n e . Unhappy : : : : : _: : :_ Happy Relaxed : : : _: : : : : Stimulated Pleased : : : : : : : : Annoyed Excited : : : : : : : : Calm Uns a t i s f i e d : : : : : : : : S a t i s f i e d Sluggish : : : : : : : : ' Frenzied Contented : : : : : : : : Melancholic J i t t e r y : : : : : : : : Dull Despairing : : : : : : : : Hopeful Sleepy : : : : : : : : Wide awake Relaxed : : : : : : : : Bored Aroused : : : : : : : : Unaroused V i v i d Autobiographical Memories 129 Consequentiality When t h i s event occurred, how much of an enduring e f f e c t did you expect i t to have on your l i f e ? no e f f e c t : : : : : : : : great e f f e c t How much of an enduring e f f e c t do you f e e l t h i s event has ac t u a l l y had on your l i f e ? no e f f e c t : : : : : : : : great e f f e c t When t h i s event occurred, how consequential did you expect i t to be to your l i f e ? very : : : : : : : : t o t a l l y consequential inconsequential How consequential has t h i s event been to your l i f e ? very : : : : : : : : t o t a l l y consequential inconsequential When t h i s event occurred, how important did you expect i t to be to your l i f e ? very t o t a l l y important : : : : : : : : unimportant How important has t h i s event been to your l i f e ? very t o t a l l y important : : : : : : : : unimportant Distinctiveness I believe t h i s event i s one that most people have never experienced. strongly strongly agree : : : : : : : : disagree I have not experienced many other events, that I consider s i m i l a r to t h i s one. strongly strongly agree : : : : : : : : disagree V i v i d Autobiographical Memories 13 0 The things about t h i s event which r e a l l y stand out i n my memory are: a) The lo c a t i o n . strongly strongly agree : : : : : : : : disagree b) The a c t i v i t i e s . strongly strongly agree : : : : : : : : disagree c) The people involved. strongly strongly agree : : : : : : : : disagree d) The emotions I experienced. strongly strongly agree : : : : : : : : disagree e) The thoughts I was thinking at the time, strongly strongly agree : : : : : : : : disagree This event was: ( c i r c l e the most appropriate response) a) the only time something l i k e t h i s occurred i n my l i f e . b) the f i r s t time something l i k e t h i s occurred i n my l i f e . c) the l a s t time something l i k e t h i s occurred i n my l i f e . d) j u s t one of a serie s of s i m i l a r events, (neither f i r s t nor last) Confidence How confident are you that t h i s event took place exactly as you remember i t ? very almost no confident : : : : : : : : confidence Surprise How surprised were you when t h i s event occurred? very s u r p r i s i n g not at a l l s u r p r i s i n g V i v i d Autobiographical Memories 131 Likelihood of occurrence Given s i m i l a r circumstances, how l i k e l y was t h i s event to occur? very very l i k e l y : : : : : : : : u n l i k e l y Cognitive Change Variables This event has changed the way I think about myself, strongly strongly agree : : : : : : : : disagree This event has changed the way I think about another person, strongly strongly agree : : : : : : : : disagree This event has changed the way I think about the world i n general. strongly strongly agree : : : : : : : : disagree V i v i d Autobiographic Memories 13 2 Appendix D Experiment 3. T-test analyses V i v i d Non-vivid p means* means* n=104 n=99 AGE NOW 21.2 20.6 ns AGE THEN 15.8 14.4 ns MEMORY AGE 5.5 6.2 ns REHEARSAL OVERT 3.0 2.6 ns REHEARSAL COVERT 4.0 3.4 <.05 PLEASANT THEN -1.1 2.4 ns AROUSAL THEN 14.9 13.4 ns CTHEN 14.6 12.1 <.01 CNOW 15.1 11.9 <.001 EE 5.6 4.6 <.01 AE 5.7 4.5 <• 01 EC 5.4 4.6 <.05 AC 5.5 4.5 <.01 EI 5.7 5.0 .05 AI 6.0 4.9 <. 01 DISTINCT TO SUBJECT 6.0 5.6 ns DISTINCT FROM OTHERS 5.7 4.9 ns DISTINCTIVE LOCATION 7.6 7.3 ns DISTINCTIVE ACTIVITIES 7.5 7.4 ns DISTINCTIVE PEOPLE 7.8 7.4 ns DISTINCTIVE EMOTIONS 7.8 6.9 <.001 DISTINCTIVE THOUGHTS 6.5 5.7 <.05 CONFIDENCE IN ACCURACY 8.1 7.2 <.001 SURPRISE 6.4 5.9 ns LIKELIHOOD 5.6 5.3 ns CHANGE SELF 4.3 4.0 ns CHANGE OTHER 4.8 4.2 ns CHANGE WORLD 4.2 4.2 ns CTHEN = consequentiality then CNOW = consequentiality now EE = expected e f f e c t AE = actual e f f e c t EC = expected consequentiality AC = actual consequentiality EI = expected importance AI = actual importance * note. Higher ratings r e f e r to greater l e v e l s of the variable or stronger agreement with the variable statement. Experiment 3. Correlational Analyses AG EN AGET M--AGE VIVID RO RC PLT ART CTHEN CNOW AGE NOW 1.00 AGE THEN .61 1.00 MEMORY AGE .17 -.68 1 .00 VIVID .05 .06 -, .04 1.00 REHEARSAL OVERT .06 .16 -, .14 .08 1.00 REHEARSAL COVERT .03 .13 -, .14 .15 .61 1.00 PLEASANT THEN -.07 .11 -, .22 -.12 .00 .10 1 .00 AROUSAL THEN -.01 .12 -. .16 .20 .07 .06 .03 1.00 CTHEN .02 .20 .24 .24 .15 .29 - .12 -.28 1.00 CNOW .16 .14 -, .03 .26 .51 .28 .08 .11 .50 1.00 EXPECTED EFFECT .07 .18 -, .17 .25 .27 .12 - .08 .25 .90 .47 1.00 ACTUAL EFFECT .15 .04 -. .09 .23 .26 .26 .04 .08 .44 .90 .51 EXPECTED CONS. .01 .18 -, .21 .19 .17 .26 .10 .22 .90 .47 .71 ACTUAL CONS. .17 .17 -, .06 .22 .27 .42 - .09 .09 .45 .92 .40 EXPECTED IMPORT. -.03 .18 -, .24 .20 .12 .27 - .14 .22 .90 .44 .71 ACTUAL IMPORT. .12 .16 -, .09 .24 .25 .53 .02 .14 .48 .91 .36 DIST. SUBJECT -.11 -.09 .00 -.02 .03 -.09 .13 .06 - .15 - .11 -.10 DIST. FROM OTHER -.07 -.02 -. .04 .02 -.03 -.09 .19 -.04 .00 - .07 -.02 DIST. LOCATION .13 .18 -, .10 .01 .04 .01 .01 .03 .10 -.03 .14 DIST. ACTIVITIES .13 .16 -. .07 .16 .04 .01 .•11 .28 .01 -.04 .07 DIST. PEOPLE -.05 -.05 .03 .08 .05 .03 .11 .05 .02 .03 .02 DIST. EMOTIONS .05 .10 -. .07 .35 .08 .16 - .03 .30 .34 .28 .30 DIST. THOUGHTS -.01 .10 .12 .26 .12 .17 - .14 .18 .23 .14 .17 CONFIDENCE -.14 -.02 -, .10 .42 .03 .07 - .09 - .01 .04 .07 .03 SURPRISE .02 -.12 .19 .08 -.02 .01 - .36 .16 .07 - .01 .10 LIKELIHOOD -.02 .03 -. .05 .00 -.08 - .01 .15 .00 .06 -.02 .06 CHANGE SELF .01 .12 .13 .10 -.23 .33 - .03 -.02 .33 .56 .29 CHANGE OTHER .16 .12 -. ,01 .14 .09 .26 - .06 .09 .26 .47 .22 CHANGE WORLD .12 .09 -. .01 .06 .16 .31 - .04 .06 .31 .54 .30 EE AGEN = age at testing AGET = age at time of event M-AGE = memory age RO = overt rehearsal RC = covert rehearsal PLT = pleasant then ART = arousal then CTHEN = composite consequentiality then scores CNOW = composite consequentiality now scores EE = expected eff e c t < < H-> c ft o cr H-O iQ 0) cr H-o S fl> 3 O in CO CO A E E C A C E I A l DSUB POTHER d L O C d A C T D P P L d E M A C T U A L E F F E C T 1.00 E X P E C T E D CONS. .37 1 . 0 0 A C T U A L CONS. .73 .48 1.00 E X P E C T E D I M P O R T . .32 .72 .36 1.00 A C T U A L I M P O R T . . .70 .44 .78 .50 1.00 D I S T . S U B J E C T .06 - . 1 6 - . 1 2 - . 1 9 - . 1 1 1.00 D I S T . FROM O T H E R .09 .02 .05 .03 .06 .20 1 . 0 0 D I S T . L O C A T I O N - . 0 6 .06 - . 0 1 .08 .01 .00 .00 1 . 0 0 D I S T . A C T I V I T I E S - .02 - . 0 5 - . 0 7 - . 0 1 .04 - . 0 7 - . 1 8 .22 1 . 0 0 D I S T . P E O P L E .03 .06 - . 0 3 .01 .01 .08 .03 .26 . 1 5 1.00 D I S T . E M O T I O N S .26 .34 - . 2 5 .27 .27 .10 . 0 1 - . 0 2 .08 .02 D I S T . T H O U G H T S .13 .22 .08 .21 .15 .03 .12 .04 - . 0 1 .09 C O N F I D E N C E .03 - . 0 1 .05 .10 . 1 1 .00 .09 .03 .07 .02 S U R P R I S E .05 .08 - . 0 5 .04 .02 .04 .09 . 0 1 .09 .00 L I K E L I H O O D .02 .01 .00 .08 - . 0 5 - . 1 2 - . 2 7 - . 1 6 - . 0 1 - . 0 4 CHANGE S E L F .48 .34 .54 .27 .50 - . 0 3 .12 .00 - . 0 8 - . 1 1 CHANGE O T H E R .42 .27 .46 .23 .42 - . 0 7 .02 .00 - . 0 1 .25 CHANGE WORLD .47 .30 .51 .25 .50 - . 0 9 .09 .08 - . 0 3 - . 0 2 A E = a c t u a l e f f e c t E C = e x p e c t e d c o n s e q u e n t i a l i t y E I = e x p e c t e d i m p o r t a n c e D S U B = d i s t i n c t t o s u b j e c t L O C = d i s t i n c t i v e l o c a t i o n d P P L = d i s t i n c t i v e p e o p l e A C = a c t u a l c o n s e q u e n t i a l i t y A l = a c t u a l i m p o r t a n c e DOTHER = d i s t i n c t f r o m o t h e r s d A C T = d i s t i n c t i v e a c t i v i t i e s d E M = d i s t i n c t i v e e m o t i o n s 1. d T H T C O N F SUR L E S E L F OTHER WORLD D I S T . T H O U G H T S 1 . 0 0 . C O N F I D E N C E .25 1 . 0 0 S U R P R I S E .05 .01 1.00 L I K E L I H O O D - . 0 6 - . 0 6 - . 2 5 1.00 CHANGE S E L F .13 - . 0 9 .03 .03 1.00 CHANGE O T H E R .06 - . 1 4 .15 - . 0 8 .41 1.00 CHANGE WORLD . 12 - . 0 6 .08 -.02 .55 .59 1 . 0 0 d T H T = d i s t i n c t i v e t h o u g h t s C O N F = c o n f i d e n c e i n m e m o r y a c c u r a c y L E = l i k e l i h o o d o f t h e e v e n t o c c u r r i n g ( n = 2 0 3 ) C r i t i c a l £ ( p = .05 .12) ( p =.01 .16) ( p = . 0 0 1 . 2 1 ) o n e t a i l e d . "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0097771"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Psychology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Characteristics underlying vivid autobiographical memories"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/28217"@en .