Baddeley's Autobiographical Memory (3rd Ed) PDF

Autobiographical memory and naturalistic approaches to memory Reading: Baddeley Chapter 11, pp. 351-362. NOTE in the syllabus it says 352-362 but obviously the chapter starts on page 351. Contents Why do we need autobiographical memory? 352 Methods of study...

Autobiographical memory and naturalistic approaches to memory Reading: Baddeley Chapter 11, pp. 351-362. NOTE in the syllabus it says 352-362 but obviously the chapter starts on page 351. Contents Why do we need autobiographical memory? 352 Methods of study 353 Theories of autobiographical memory 360 Emotion and autobiographical memory 362 Variations in autobiographical memory function 369 Neural basis of autobiographical memory 380 Concluding remarks 383 Summary384 Points for discussion 386 Further reading 386 References386  C h a p t e r  11 A U T O B I O G RAPH I C A L M E M O R Y Michael C. Anderson y new significant other and I had fin- after the film, as though it had been planned M ished cooking dinner and we had arranged the table, with candles and all along. Outside the theater, I frantically prepared my lecture for the next 60 minutes. pleasant music. As we raised our wine glasses Now that I look back on this event years for a toast, my mobile phone rang. Reluc- later, I can laugh at the irony of somebody tantly, I answered the call, which was from who studies forgetting for a living, forgetting my Ph.D. student, Justin Hulbert. Justin said, to give a lecture on a film about wanting to “Are you almost here?” in a polite, if tense forget. In truth, I actually rather enjoy telling tone. Confused, I said “What do you mean? the story, as it fits a well-­justified theme about Where? I am just sitting down to eat.” me as an absent-­minded professor. If I ever Awkward silence ensued. Finally, he said write an autobiography, I will surely try to fit “Eh, you are supposed to be downtown with this story in. We all have a collection of such me in the cinema. Remember, you agreed to personal memories that are special in that give an introductory lecture to the film they are part of our life stories, and so seem Eternal Sunshine of the Spotless Mind, as more than just ordinary episodic memories. part of the Memory Film Festival.” The Autobiographical memory refers to the mem- Memory Film Festival was the brainchild of ories that we hold regarding ourselves and Amy Milton, a professor of psychology, our interactions with the world around us, organized as part of the weeklong Cambridge that help to define who we were at different Science Festival, a popular annual event at times in the past, who we are currently, and the University of Cambridge. Apparently, I who we hope to be in the future. It includes had agreed to give this lecture because of my not only episodic memories that form part of own research on forgetting, and I had, poeti- our life stories, but also semantic autobio- cally, completely forgotten. Pressing ahead, graphical memory that includes historical Justin said “There are easily 200 people in facts, traits, and knowledge states that are not the theater now, expecting the lecture in 15 unique to any particular place or time. minutes.” This news elicited the feeling that Autobiographical memory is clearly one has during a dream, when one discovers important to each of us, as it helps to define oneself transported back to a busy high school hallway, without clothes. Shocked, I apologized to my significant other, immedi- KEY TERM ately leaving her befuddled and alone at the table, as I sped to the theater. The organizers Autobiographical memory: Memory across the lifespan for both specific events and self-­ cleverly told the audience that they had related information. decided the lecture would be better given 352 Autobiographical memory our sense of ourselves as people. But is auto- can be a very pleasant and socially supportive biographical memory a separate kind of activity (Neisser, 1988). In my own case, memory? On the one hand, it almost cer- getting together with friends and reminiscing tainly depends on the episodic and semantic about our fun times is an example. Con- memory systems we have already discussed, versely, when autobiographical memory is and so may not be qualitatively different. disrupted by amnesia or dementia, this can be Remembering facts about ourselves, such as one factor that impairs relationships (Robin- our name, when we went to school, and son & Swanson, 1990), leading to the feeling where we live, is autobiographical but forms that “This is not the person I married.” Auto- a personal aspect of semantic memory. biographical memories can also play an Remembering what you had for breakfast important role in creating and maintaining today is also autobiographical but involves our self-­representation, hence the value of recollecting a specific episodic experience. reminiscence therapy (Woods, Spector, Jones, The fact that autobiographical memory Orrell, & Davies, 2005), a process described involves both of these suggests that it may be in Chapter 16 whereby elderly patients with a complex blend of these other forms of memory problems are encouraged to build up memory. Nevertheless, as we will see later on a set of reminders of their earlier life based in this chapter, evidence from neuroscience on photographs and personal mementos­ —­ suggests that the idea that autobiographical items that bring back memories of their memory is simply a mixture of episodic and younger days. Finally, autobiographical rec- semantic memory may not be entirely ade- ollection can be used for emotion regulation, quate, as it appears to involve qualitatively as when we need to cope with adversity, or distinct brain mechanisms. Even without con- build confidence. One of the problems of sidering the mechanisms involved, it is worth depression is that patients find it difficult to distinguishing autobiographical memory recollect positive life experiences when simply because the role that it plays in our depressed, whereas negative recollections are lives differs in interesting and important ways more readily available, a retrieval effect from other functions of memory. known as mood-­congruent memory, which is We will begin by discussing the function one form of context-­dependent memory dis- of autobiographical memory and why it is cussed in more detail in Chapter 8. Healthy important, leading on to the thorny question individuals often engage in nostalgia about of how to study it. The problem here is that, times past to maintain intimacy with unlike most of the research we have discussed friends, to teach others, and to enhance self-­ so far, the experimenter typically has no perceptions (Cheung, Wildschut, & Sedikides, control over the learning situation, which 2018). makes it difficult to analyze the processes However, although these functions might involved in either the acquisition or forget- be plausible, they are largely speculative. In ting of autobiographical memories. an attempt to obtain empirical evidence on this matter, Hyman and Faries (1992) ques- tioned people about memories they fre- quently talked about, and the situations in WHY DO WE NEED which they were discussed. They found very AUTOBIOGRAPHICAL few reports of autobiographical memory being used directively to solve problems, with MEMORY? the sharing of experience and passing on of advice being more common. In a subsequent Williams, Conway, and Cohen (2008) study, they used cue words to prompt memo- propose four functions of autobiographical ries, finding a distinction between memories memory. These include directive functions, that were used internally for self-­related func- for example what happened the last time you tions and those used in interacting with tried to change a car tire, and a more social others, but again little evidence of directive function; sharing autobiographical memories use of autobiographical memory.  Autobiographical memory 353 Bluck, Alea, Habermas, and Rubin (2005) Diary studies devised the Thinking About Life Experiences (TALE) questionnaire, specifying particular A central problem in studying autobiographi- situations and then categorizing the resulting cal memory is that of knowing what was ini- reports as: directive, self-­related, nurturing tially experienced. Whereas laboratory existing social relationships, or developing studies can carefully control which stimuli new social relationships. The factor analysis participants are exposed to and also the of the results found considerable overlap encoding conditions, in autobiographical between the directive function, the self-­related memory studies, every person’s memories will function, and those related to nurturing and differ, and experimenters have no way of developing relationships (e.g., I enjoyed talking knowing whether recall is accurate. One solu- to John; so I think I’ll accept his party invita- tion to this is to record events in a diary that tion). Hence, although it remains plausible allows later memories to be objectively that autobiographical memory has a number checked. Linton (1975) used this method to of different functions, it is unclear that they study her own autobiographical memory. She are clearly separable into different categories kept a diary for over five years, recording two in actual practice. events per day, each being briefly described One weakness with the research described and written on an index card. She tested so far is the problem of adequate methodol- herself each month by randomly picking out ogy. The studies assume, for example, that two index cards and deciding whether she participants are aware of the function of such could remember the order in which incidents memories and can remember their autobio- occurred and the date. Because she chose graphical memories and the situations that cards at random and then replaced them, she evoked them in sufficient detail to categorize would sometimes test herself on the same them. In an area as complex as autobiograph- incident on several occasions. As Figure 11.1 ical memory, there is clearly a need for the shows, she observed a powerful effect: The development of a range of methods of study. more often an event was probed, the better it I discuss this next. was retained. This provided further evidence for the value of retrieval practice in long-­term learning as discussed in Chapter 5. A classic diary study was carried out by METHODS OF STUDY the Dutch psychologist Willem Wagenaar (1986), who kept a diary for over six years, One method of tackling this problem is to use on each day recording two events, together diaries in which participants record events, with four features or cues to that event. As and subsequently try to remember them. This shown in Figure 11.2, he recorded who was is a useful approach but one that places involved, what the event was, where it onerous and persistent demands on particip- occurred, and when. He also rated the inci- ants. A second approach is to probe memory, dent for its saliency and whether it was some- for example asking for a memory associated thing that happened frequently or was rather with a cue word such as river, then analyzing unusual, in addition to recording the degree the nature of the responses. A third method is of emotional involvement and whether this to ask for memories associated with either a was pleasant or unpleasant. He recorded a specific time period, or a major public event total of 2,400 incidents. He then tested his such as the 9/11 attack on New York. memory by selecting an incident at random Finally, as in the case of semantic and epi- and cuing himself with one, two, or three sodic memory, we can learn a good deal from retrieval cues, randomizing the order in which individual differences in autobiographical the who, what, where, and when cues were memory as well as what happens when auto- presented. Figure 11.3 shows the mean per- biographical memory breaks down, as the centage of questions answered correctly as a result either of brain damage or emotional function of number of cues. As you can see in stress. These approaches are discussed next. this figure, his recall improved as he added 354 Autobiographical memory 70 60 1st test 50 2nd test Percent forgotten 40 30 3rd test 20 Figure 11.1 Probability of forgetting an 10 4th or later test autobiographical diary item as a function of elapsed time and number of prior tests. 0 From Linton (1975). 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Copyright W.H. Freeman. Age of item (years) Reproduced with permission. more cues, an effect we discussed in Chapter Supper, accompanied by scientific colleagues. 8 as a key determinant of retrieval success. The process of selecting the event would in Moreover, Wagenaar found that the who, itself involve retrieval, and in effect a what, and where cues tended to be equally rehearsal, while the process of deciding on his good at evoking a memory, whereas the when who, where, what, and when cues would cue, which simply provided the date, was involve a relatively deep level of processing much less efficient. This is perhaps not sur- (Craik & Lockhart, 1972). This degree of prising. Can you remember where you were selection and implicit rehearsal is a problem on July 19 last year? Neither can I, although for diary studies, because they result in mem- as we shall see later, some people can. ories that are atypically well encoded. Wagenaar reports that he found the task A somewhat more naturalistic approach to be surprisingly difficult and unpleasant, to encoding of autobiographical memories is but that given sufficient cues he could recol- to use events reported in letters. One of the lect most of the incidents eventually. In a authors of this text, Alan Baddeley, has used number of cases, he could not remember any- this method, based on a series of letters sent thing, despite all his recorded cues. However, to his widowed mother during a year in Cali- in those cases where another person was fornia some 40 years ago (Baddeley, 2012). involved, they would typically be able to Baddeley went through the letters, identifying evoke a recollection, which could be verified anything that could be regarded as an integ- by his providing additional information. rated episode, then classifying each on the Does that mean that we never forget any- basis of the extent to which he could thing? Almost certainly not. Wagenaar remember it. He distinguished three degrees selected events that were most likely to be of vividness of the recollection, together with highly memorable; for example, going to see episodes that “he knew” had happened but Leonardo da Vinci’s painting of The Last of which he had no recollective experience,  Autobiographical memory 355 Figure 11.2 An example of a recorded event from Wagenaar’s diary study (1986). Reproduced with permission from Elsevier. and those that were completely forgotten. Of landscape of potential memories extending 62 episodes identified in the letters, Baddeley into the distance with striking peaks of vivid judged that he could remember 23, about memories and less clear valleys, the experi- half of these vividly. One example that stood ence seemed to him much more analogous to out for him involved losing his voice on a lec- perceiving a limited series of islands of turing trip, then having a lively dinner with memory in a sea of forgetting. Furthermore, Endel Tulving, and the frustration of being the fact that the “islands” appear to depend unable to croak his own opinions with ade- on retelling over the years implies that they quate vigor. The vivid memories were of non- themselves may not be true memories, but trivial unique events that he remembered rather memories of memories, a rather sober- having told others about. Of the 62 total ing thought! In fact, the tendency for our recorded episodes, Baddeley had forgotten 26 autobiographical recollections to be overly completely, nearly all trivial, with no recol- populated with memories about which we lection of retelling. Retelling of course is a have frequently told stories illustrates how form of rehearsal that, as Linton’s (1975) autobiographical memory may be different diary indicated, has a major effect on sub- from episodic memory, as it is conventionally sequent recall. studied in the laboratory. The experience of rereading his letters But perhaps this is only true of distant changed Baddeley’s view of his own autobio- memories? Why should we want to remember graphical memory. Instead of seeing it as a relatively trivial events that happened 40 356 Autobiographical memory the act of providing the event features in the 100 Retention judgment tape recorder to serve as subsequent cues would again have involved atypically deep encoding, perhaps explaining the minimal 80 Critical details recall that did occur. A more detailed analysis of the nature of the items recalled was made by Conway, Percent retention 60 3 cues Collins, Gathercole, and Anderson (1996) in a study involving two participants who kept diaries over a period of months, recording 40 2 cues both “events” and “thoughts.” These were then mixed with plausibly invented altern- 20 atives and recognition was required. This was 1 cue question followed by a categorization as to whether the item was “remembered,” meaning that 0 recognition was accompanied by a feeling of 1 2 3 4 5 recollecting the initial experience, or simply Retention period (years) “known” (see Chapter 8 for discussion of this distinction). True events were more likely to evoke a remember response than invented Figure 11.3 Recall of autobiographical incidents as a but plausible foils, with items classified as function of time, given one, two, or three retrieval cues. Reproduced with permission from Elsevier “events” being twice as likely to evoke recol- lection as entries that were “thoughts.” The reduced feeling of remembering thoughts is years ago? And perhaps it is not surprising interesting in that it points to an important that surviving memories are rather special, or role of perceptual experience in anchoring the at least worth telling others about. Would experience of remembering, a topic to which nonselected relatively trivial memories we shall return later. survive over a much shorter period than In conclusion, diary studies have been those reported in diary studies? Brewer useful in giving some idea of the nature of (1988) tried to avoid the biased selection of autobiographical memory, and of the relative recorded memories in a study that sampled importance of different types of events and events at random. His 10 participants were experiences. They do, however, suffer from each given a beeper and a tape recorder. The problems of sampling bias in the events beeper went off at random intervals, at which recorded, together with a tendency for the point participants were to say what they were event reporting process itself to result in the doing, where, what the significance of their enhanced learning of the events selected. activity was, its goal-­directedness, and their Finally, the method requires considerable per- emotional state. The incidents were tested at severance from the diarists, who are therefore delays ranging from 0 to 46 days, using one likely to be a small and atypical sample of the or other of their ratings as a cue. A total of general population. 414 events were recorded. When subse- quently tested, 26% were correctly recalled, 28% were wrong, and 46% evoked a blank. The memory probe method It is likely that, given more cues, more would be recalled but it seems very unlikely that all An alternative to the diary method is that of of the 74% failed memories would be recol- cued recall, a method first used by Galton lected. Indeed, if the study by Misra, (1879). It was subsequently revived by Marconi, Peterson, & Kreiman (2018) dis- Crovitz and Shiffman (1974), who gave their cussed in Chapter 5 on learning is any guide, participants a word and asked them to recol- it is astonishing participants recalled any of lect an autobiographical memory associated the events at all. Note that in Brewer’s study, with that word. For example, given the cue  Autobiographical memory 357 word horse, this might evoke a memory of the first time you rode a horse. The method has also been adapted to probe for memories from a given time period such as childhood, or of a particular type of incident, for example a happy memory. Despite its simpli- city and relative lack of control, this method has been used widely, and productively, including in many studies of brain imaging, as we will discuss later on. A prominent feature of probed autobio- graphical memories is their distribution across the lifespan. When left free to recall memories from any period in their life, all healthy participants, whether young or old, tend to recall few autobiographical memories from the first five years of life, termed infantile amnesia (see Chapters 9 and 14 for further discussion). They also tend to produce plenty of memories from the most recent period. Those over the age of 40, however, also show a marked increase of memories from the period between the ages of 15 and 30, the so-­called reminiscence bump (Rubin, Wetzler, & Nebes, 1986). A cross-­cultural study illustrated in Figure 11.4 shows a similar pattern across participants Sir Francis Galton (1822–1911), a Victorian from China, Japan, Bangladesh, England, and polymath, who in addition to his classic study of autobiographical memory, was a tropical explorer, geographer, meteorologist, anthropologist, and KEY TERM statistician. Reminiscence bump: A tendency in participants over 40 to show a high rate of recollecting personal experiences from their late teens and early twenties. 35 Japan Bangladesh 30 UK Percent of memories China 25 US 20 All 15 10 Figure 11.4 Lifespan retrieval curves for 5 participants from five 0 countries. From Conway et 5 10 15 20 25 30 35 40 45 50 55 60 al. (2005). Copyright © 2005 Age at encoding (in 5-year bins) Sage Publications. Reprinted by permission. 358 Autobiographical memory the US (Conway, Wang, Hanyu, & Haque, the basis of memories and experiences (see 2005). However, there are cultural differ- Chapters 9 and 14 for more on infantile ences in the average date for the first amnesia). memory, which occurs at an average age of Most interpretations of the reminiscence 3.8 for the US and 5.4 for Chinese particip- bump tend to focus on the fact that this is a ants (Wang, 2006a, 2006b). This might period when many important things in our reflect differences in the way that mothers lives tend to happen. Berntsen and Rubin talk to their children, with the US interaction (2004) asked their participants to rate a tending to be more elaborate, emotionally number of important life events, finding that oriented, and focused on the past than occurs the average age for first falling in love was 16 in Chinese culture (Leichtman, Wang, & Pil- years, college memories tended to be a rather lemer, 2003). This might also account for a later 22 years, marriage at an average age of tendency for US recollections of early memo- 27, and children at 28. All fell within the ries to be longer, more elaborate, and more period of the bump, making this an emotionally toned and self-­ focused than important period within what is sometimes occurs with Chinese respondents, whose rec- known as the life narrative. This represents a ollections tend to be briefer and to have a coherent account that we create for ourselves stronger collective than individual emphasis as we progress through life­—­the story of who (Wang, 2001). we are and how we got to this point in our There have been a number of attempts to life. Events that influence this are likely to be explain the pattern of autobiographical mem- important to us, to be more likely to be ories across the lifespan. It probably reflects retrieved, and to be more deeply encoded. both a recency effect (see Chapter 3, p. 50) Indeed, the attachment of salient events such and at least two other processes, one as these to an organized narrative schema in accounting for infantile amnesia (the lack of long-­ term memory likely confers powerful memories from the first one or two years of benefits to both consolidation and memory life) and the other concerned with the high search, as discussed in Chapter 6 on episodic rate of recalling episodes from the teens and memory. Furthermore, such events as begin- twenties. Many interpretations of infantile ning college, making new friends, and falling amnesia have been proposed. Early theoret- in love are all likely to be emotionally ical proposals to explain this have included intense, a factor that increases the accessibil- explanations based on Freudian repression ity of memories (Dolcos, LaBar, & Cabeza, and the late development of the hippo- 2005), particularly when these are positive campus. More recently, however, a compel- and occur in young adulthood (Berntsen & ling account of infantile amnesia attributes Rubin, 2002). this striking phenomenon to the rapid rate of Glück and Bluck (2007) further elaborate increase in new neurons in the hippocampus, the life narrative hypothesis. They collected a the introduction of which into the neural total of 3,541 life events from 659 particip- network makes it harder to reinstate memo- ants aged between 50 and 90 years. Particip- ries prior to their inclusion. This phenom- ants were asked to rate their memories on enon is known as neurogenesis-­ induced emotional valence, their personal importance, forgetting (see Chapter 9 for a discussion of this process). This account has the benefit that it explains infantile amnesia in many KEY TERM different species of nonhuman animals. However, given the powerful role of schemas Life narrative: A coherent and integrated in enhancing consolidation of long-­term epi- account of one’s life that is claimed to form the sodic memories, discussed in Chapter 6, the basis of autobiographical memory retrieval. A life lack of early episodic memories may also be narrative provides an organized set of schemas with which key episodic events can be integrated, related to the absence, during infancy, of a both increasing the chances of consolidation, and coherent concept of self, a general set of making memory retrieval efficient. schemas that would gradually be built up on  Autobiographical memory 359 and the absence of a reminiscence bump for negative autobiographical memories illustrate the role that motivation plays in shaping which memories remain accessible in long-­ term memory, a topic discussed in Chapter 10 on motivated forgetting. An intriguing exception to the reminis- cence bump in a person’s early twenties occurs when memories are cued by smell. Despite an initial report by Rubin, Groth, and Goldsmith (1984), of equivalence across verbal, visual, and olfactory cues, Chu and The reminiscence bump occurs in early adulthood Downes (2002) found that memories evoked and reflects memories from a period when many by smell peaked at an earlier age (6–10 years) important life events, such as falling in love, getting than the memories found in the typical ver- married, and having children, tend to happen. bally cued reminiscence bump. Willander and Larsson (2006) replicated this using a sample of 93 volunteers ranging in age from 65 to 80 and the extent to which the rememberer felt years. They cued with items that could not that they had control over events. A reminis- only be represented as a word, but also as a cence bump was found, but only for positive picture or a smell (e.g., violet, tobacco, soap, events over which participants felt that they whiskey). Like Chu and Downes, they found had a high degree of control, a result that a distinct tendency for smells to evoke memo- they interpret as consistent with the import- ries that are rated by their participants as ance of autobiographical memory in creating earlier than visually or verbally cued events. a positive life narrative (Figure 11.5). In addi- How could we explain this? Are odor-­ tion, as can be seen in Figure 11.5, the overall induced memories more emotional? Both rate at which people recall positively memo- Herz (2004) and Willander and Larsson ries overwhelmingly outstrips both negative (2007) found that they were. and neutral memories, despite the fact that, It is, of course, the case that the probe in the weeks and months following initial studies described all depend to some extent encoding, negative events tend to be unusu- on the accuracy with which participants can ally memorable. This positivity bias is a widely date events. As we saw from Wagenaar’s reported phenomenon, as we will discuss diary study, memory for dating of an incident later in this chapter. Both the positivity bias was the weakest of all the cues. This also 21 Positive 18 Neutral Negative Percent of memories 15 12 Figure 11.5 Distribution 9 of involuntary memories for participants who were over 6 40 years old. Only positive memories show the 3 reminiscence bump. From Glück and Bluck, (2007). 0 Copyright © The 0 10 20 30 40 50 Psychonomic Society. Age at time of event Reproduced with permission. 360 Autobiographical memory presents a problem for the many practically memory, such as seeing the Eiffel Tower in the oriented survey studies that are retrospective rain. Thus, autobiographical memory includes in nature, requiring respondents to both generalized knowledge of events, and spe- remember, for example, when they last went cific episodes. Later in the chapter, we will to the doctor, or how often they eat certain discuss special cases of people who can recall types of foods. A study by Means, Mingay, the former type of knowledge, but who Nigam, and Zarrow (1988) asked patients have severe deficits in the latter aspect of who had made at least four medical visits in re-­experiencing. the last year to recall and report them, subse- Such autobiographical recollections are quently checking against the doctor’s records. transitory and are constructed dynamically Performance was poor, particularly for visits on the basis of the autobiographical know- that had clustered (25% correct versus 60% ledge base. The knowledge base itself ranges for more isolated occasions). People tend to from very broad-­brush representations of life- date events indirectly, either by recollecting time periods to sensory–perceptual episodes, incidental features such as the weather or which are rapidly lost. Finally, the whole “the trees were bare,” or by linking it to system depends on the interaction between some other event that can itself be dated, the knowledge base and the working self. The such as a holiday in Paris or the eruption of working self is assumed to play a similar role Mount St. Helens (Baddeley, Lewis, & in autobiographical memory to that played Nimmo-­ Smith, 1978; Loftus & Marburger, by working memory in cognition more gener- 1983). These, in turn, are likely to be located ally (Conway & Pleydell-­ Pearce, 2000). within the broader context of a life narrative. These broad ideas were developed by Conway (2005) into a more detailed account of the way in which the self interacts with memory (Figure 11.6). THEORIES OF The working self comprises a complex set AUTOBIOGRAPHICAL of active goals and self-­images. For example, I have the active goal of describing Conway’s MEMORY ideas and am doing so while on a train en route to London’s Kings Cross railway The preceding discussion focuses on how auto- station. The goals active in the working self biographical memory is measured, and some modulates access to long-­term memory and is interesting generalizations discovered with itself influenced by LTM. To write this, I those methods. But how does autobiographical need to access my knowledge of Martin Con- memory work? What processes does it involve? way’s views. The working self comprises both One attempt to develop an overall theory of conceptual self-­knowledge­—­my occupation, autobiographical memory is that proposed by my family background, and my professional Martin Conway (2005). Conway defines auto- aims­—­which in turn are socially constructed biographical memory as a system that retains on the basis of my family background, the knowledge concerning the experienced self, the influence of peers, school, myths, and other “me.” Autobiographical memory is always factors that make up the complex representa- accessed by the cues about the content of the tion of myself. memory desired, but the results of memory search do not always produce recollective experience; hence you might know that you KEY TERM had a trip to Paris last year, but only recollect Autobiographical knowledge base: Facts the episodic detail later, or indeed not at all. about ourselves and our past that form the basis Such recollective experiences occur when auto- for autobiographical memory. biographical knowledge retains access to asso- Working self: A concept proposed by Conway ciated episodic memories with perceptual to account for the way in which autobiographical details, for example when the knowledge that knowledge is accumulated and used. you went to Paris connects with a specific  Autobiographical memory 361 Life story Themes Work theme Relationship theme Lifetime periods Working at Friends The conceptual self university “X” with “Y” Others Others Activities Activities Locations Locations Projects Projects Goals Goals General events Prof. Smith Psych. Dept. Dept. talks Grant “Z” Promotion Episodic memories Figure 11.6 The knowledge structures within autobiographical memory, as proposed by Conway (2005). Copyright © Elsevier. Reproduced with permission. To summarize, the working self is a The autobiographical knowledge base is complex knowledge structure that con- proposed to have a hierarchical structure, tributes to encoding information about what with an overall life story being linked to sub- is, what has been, and what can be. To be sidiary themes; work and personal relation- effective, however, it needs to be both coher- ships, for example. These in turn split up into ent and to correspond reasonably closely different time periods: for example, When I with outside reality. When this link is lost, was an undergraduate; My first job; My problems occur, which might­ —­in extreme hopes for the future. These comprise a number cases­—­lead to confabulation or delusion (see of general “events,” which can include indi- Chapter 16). Conway and Tacchi (1996), for viduals and institutions as well as activities: example, describe a patient suffering frontal-­ for example, The psychology department; lobe damage following a road traffic accident Professor Smith; Departmental talks; Promo- who had comforting but totally false memo- tion. These are still conceptualized at a relat- ries of the support provided by his family. ively abstract level but can lead to specific 362 Autobiographical memory episodic memories; for example, my inter- memory are shaped by the surrounding view with Professor Smith on applying for a sociocultural world, are not depicted in job, or the last departmental research talk I Figure 11.6. heard. These in turn might have been stored Conway and Jobson stress that the opera- at a more fundamental level containing more tion of autobiographical memory is goal-­ detailed sensory-­ perceptual information; for related, and that different goals tend to example, the room where the interview was predominate in different cultures. There is, for held, the weather outside, or Professor example, considerable evidence for differences Smith’s tone of voice in offering me the job. between individualistic cultures that tend to In recollecting an event, it is this essentially dominate in the West, and more communal arbitrary sensory detail that typically con- and collective attitudes that are characteristic vinces us that we have a genuine memory of many Asian cultures. Wang (2008) studied rather than a confabulation (Johnson, Foley, autobiographical memory in US participants Suengas, & Raye, 1988). Such detail is often who come from an Asian background, visual in character, which is one reason why probing autobiographical memory related to vivid visual flashbacks are so convincing and their US or Asian identity and finding the potentially so disturbing. latter generated more socially oriented memo- Following Tulving (1989), Conway refers ries in contrast to the greater self-­focus for US-­ to the process of recollecting such detail and related memories. Marian and Kaushanskaya experiencing it as being based on auto- re-­ (2004) interviewed Russian-­ English bilingual noetic consciousness, the capacity to reflect participants in both languages. Regardless of on our thoughts. This ability to reflect on our the language of encoding of a specific autobio- memories is of course essential in deciding graphical memory, recollections in Russian whether a recollection is an accurate record were more collectivist than those in English. of our past or a confabulation. Accessing At a more extended and integrated level, life such detailed features tends to be relatively schemas are also influenced by culture; hence a slow, typically taking several seconds, celebration such as a Bar Mitzvah is likely to whereas access to semantic memory is often tie participants into traditional Jewish culture performed almost immediately (Haque & and a first communion into a Catholic context Conway, 2001). As we saw in Chapters 4 (Berntsen & Rubin, 2004). and 8, patients with frontal-­lobe damage can Conway’s theory provides a useful have difficulty both in accessing autobio- framework that pulls together what we know graphical memories and also, once accessed, about autobiographical memory, which in in evaluating them. turn is likely to lead to further more theor- In a later account of the model, Conway etically oriented questions. For example, how and Jobson (2012) discuss the role in autobi- might we test the assumption that the autobi- ographical memory of factors such as the ographical database is divided in the way parents’ reminiscing style and subsequent dis- proposed by Conway (2005)? cussion of memories with peers, as well as broader social factors such as the manner of Required reading for 2023 ends right here thought in the local culture, local heroes and role models, myths, and general attitudes as EMOTION AND to what is good and right. Such influences, in AUTOBIOGRAPHICAL which the inner processes of autobiographical MEMORY KEY TERM So which memories from your life will you remember when you are 80? What factors Autonoetic consciousness: A term proposed determine which experiences “stick?” Natur- by Tulving for self-­awareness, allowing the ally, one answer to this has to do with the rememberer to reflect on the contents of episodic role that memories play in our life narratives, memory. and the corresponding integration this will  Autobiographical memory 363 cause with long-­ term memory schemata. Another factor concerns how often we recall and revisit a memory, as discussed earlier through the diary study reported by Marigold Linton. Apart from these factors, however, one must also consider the role of emotion in shaping what we remember. Are we destined to permanently remember anything that is highly emotional, regardless of whether it is pleasant or unpleasant? Does the emotion attached to a memory persist stubbornly over time, or does the punch that a memory has decline or event transform? Using some of the methods described at the outset, research on autobiographical memory has addressed these important questions, sometimes with Flashbulb memories are typically vivid, clear and surprising results. We discuss several key persistent. What were you doing when you heard phenomena next. about the World Trade Center attacks on September 11th 2001? Flashbulb memory event and its physical context. In subsequent One might imagine that extraordinary and years, this has proved to be an extremely emotionally significant events could have a popular area of study. It now seems that special privileged status in your autobio- whenever a disaster occurs, a cognitive psy- graphical memory. Do you remember where chologist somewhere will be devising a ques- you were when you first heard of the 9/11 tionnaire to establish whether flashbulb attack on the World Trade Center? Unlike memories have occurred and trying to answer humdrum events such as routine visits to the some of the questions raised by Brown and doctor, certain occasions appear to give rise Kulik’s claim. to remarkably clear detailed and persistent There is no doubt that people do report memories. Brown and Kulik (1977) asked very vivid recollections of the point at which people to recall how and when they had first they remember hearing about major disasters. heard of the assassination of President It is also the case that the probability of Kennedy. They found a degree of vividness report of a flashbulb memory depends on the and detail that was surprising, leading them degree to which the rememberer was likely to to propose a new kind of memory system, be affected by the event. African-­Americans which they termed flashbulb memory. They were more likely to have a flashbulb memory argued for a separate process that, given concerning the deaths of Martin Luther King appropriate conditions, leads to a special and Malcolm X than were European Ameri- mechanism resulting in a qualitatively can participants (McCloskey, Wible, & different memory record. They termed this Cohen, 1988), and Danes who reported an process the “now print” mechanism, whereby involvement with the Danish resistance extreme emotion was assumed to lead to an movement were more likely to have a flash- almost photographic representation of the bulb experience, and be able to report on the weather, time of day, and day of the week for the invasion and liberation of Denmark than KEY TERM did those who were less directly involved (Berntsen & Thomsen, 2005). Flashbulb memory: Term applied to the But do we need to assume a special detailed, vivid and apparently highly accurate mechanism to account for these results? memory of a dramatic experience. The Brown and Kulik conclusions have been 364 Autobiographical memory scrutinized on two fronts. The first concerns way is much easier for the experimenter than the question of whether flashbulb memories providing adequate cues for an everyday are as accurate as they seem, and the second event in someone else’s life. The latter study, concerns whether one needs a special mech- therefore, might suggest that when a carefully anism to explain them. In a study based on crafted control memory is used, there is little the Challenger space disaster, Neisser and difference in the rate with which flashbulb Harsch (1992) compared the recall of the and everyday memories are forgotten over experience of learning about the event, time. Nevertheless, Talarico and Rubin did testing people after one day and retesting find that, despite the comparable retention after 2½ years, finding a substantial drop loss for details, participants reported higher in accuracy. For example, after one day, ratings of conscious recollection, vividness, 21% reported first hearing about the disaster and other phenomenological aspects of the on TV, whereas after 2½ years this had memories for flashbulb memories, consistent increased to 45%. Similarly, Schmoick, with superior retention. As we will discuss Buffalo, and Squire (2000) reported consider- later on, the phenomenological experiences able forgetting of hearing the result of the OJ of remembering (perceived vividness and Simpson trial over a period of 32 months. imagery) play a critical role in the function of A further problem is the question of autobiographical memory. Indeed, some what should be the baseline against which individuals with severely impaired autobio- one judges whether a memory is unusually graphical memory, can remember the facts of accurate or vivid. Rubin and Kozin (1984) an event whilst having no ability to re-­ report that memories of high-­ school gradu- experience it in their mind’s eye. Talarico and ation or of an early emotional experience Rubin’s findings therefore suggest that the can be just as clear and vivid. Should one critical feature defining flashbulb memories compare proposed flashbulb memories may not be the extent or accuracy of factual against distinctive, but nonlifethreatening detail, but the preservation of the sensory events? Or against everyday events? Does the experiences of details that are remembered, choice of control memories and how to and their capacity to evoke reliving. measure them influence one’s conclusion Although flashbulb memories may not be about whether flashbulb memories are unusu- as impressively accurate and persistent as ally well retained and vivid? suggested by Brown and Kulik, there is no To illustrate this issue, consider a study doubt that people do have vivid autobio- by Davidson, Cook, and Glisky (2006), who graphical memories of flashbulb incidents. contrasted memory for the 9/11 World Trade There are a number of reasons why this Center attack with everyday memories, might be. First, such incidents are highly dis- finding that after a year there was a correl- tinctive, with little danger of their being con- ation of 0.77 between the initial and sub- fused with other events, which is not the case sequent recollection for the 9/11 incident, for most everyday memories. Second, we tend indicating very good retention, compared to talk about such events and watch them with a correlation of only 0.33 for more repeatedly on TV; in effect, rehearsing them. everyday memories. In contrast, however, Third, they tend to be important events that Talarico and Rubin (2003) found the same potentially change some aspect of our lives degree of loss of detail of 9/11 memories and and surroundings; and fourth, they tend to everyday memories. The crucial difference give rise to heightened emotions. Given that between these two studies appears to be that, all of these factors are likely to enhance whereas Talarico and Rubin’s participants memory in one way or another, do we need themselves produced and recorded their an additional quite separate theory? The everyday events, (and hence generated their debate on this point continues in the field (in own retrieval cues) in the Davidson et al. our later discussion of PTSD, for example, study, the experimenters chose the events to we will return to this issue); indeed, despite be recalled by participants. Cuing an excep- initial skepticism about Brown and Kulik’s tional event (like 9/11) in an unambiguous proposal, considerable evidence is consistent  Autobiographical memory 365 with superior encoding (see, e.g., Conway, hour was up, he asked them to recall 12 2013 for a thorough discussion). One might memories while they were still floating, and ask whether it is worthwhile attempting to rate each for pleasantness. Participants rated untangle these various contributions that their recalled events as being pleasant (66%) operate under conditions that are by their more often than unpleasant (33%). This very nature hard to control? But no doubt effect does not reflect a strategic bias in how studies on this topic will continue, not only people voluntarily recall their memories. For because the phenomenon is dramatic and example, Berntsen (1996) asked participants intriguing, but also because the answers to keep a diary of memories that involuntar- matters to victims of trauma, as we discuss ily popped into mind, noting each memory later on. It is surely helpful, however, to as soon as it occurred. Such memories are attempt to understand the possible contribu- generally triggered by stimuli in the tions independently, perhaps subsequently immediate environment, and occur automati- attempting to bring them to bear on the phe- cally, and so are unlikely to reflect voluntary nomenon of flashbulb memory. memory search. Here too, Berntsen’s parti- cipants were positively biased, reporting that 49% of the events were positive, 32% were Positivity bias neutral, and 19% were negative. Indeed, earlier in this chapter, we reported a vivid From the foregoing discussion, one might example of this positivity bias (see Figure assume that emotional memories may, in 11.5) in the context of involuntary memory general, enjoy a persisting advantage in long-­ retrieval. In that example, the positivity bias term autobiographical memory, and be dis- seems to grow with the age of the memory, proportionately represented, relative to more suggesting a process that gradually renders mundane and neutral events. On the whole, negative memories less accessible. Even when emotional events often are retained better. neutral probe cues are used (e.g., pool, medi- There is, however, a counterintuitive finding cine) and participants are explicitly asked to in research on autobiographical memory that recall either a positive or a negative memory suggests that negative memories, over time, and given a full 25 seconds per cue, they are do not fare nearly as well as do positive reliably less able to recall negative memories memories for most people. The dispropor- (35%) compared to positive memories tionate accessibility of positive, relative to (43%) (Storm & Jobe, 2012). This shows negative autobiographical memories, over that the bias is not merely a tendency to time is referred to as the positivity bias. favor reporting positive memories over neg- One of the first studies to examine this ative ones, but instead reflects difficulty in issue was reported over 70 years ago (Wald- accessing them. fogel, 1948). Waldfogel gave participants 85 What could account for this clear bias in minutes to write down all of the memories accessibility? One possibility is that people that they could remember from the first eight are more motivated to selectively remember years of their childhood, and rate them as positive events in their lives, and to forget the pleasant, unpleasant, or neutral. Waldfogel’s negative events, which generate unpleasant participants rated 50% of their events as feelings about themselves (see Chapter 10 on pleasant, 30% as unpleasant, and 20% as motivated forgetting). The previously men- neutral. One explanation is that stimuli in tioned study by Storm and Jobe provided their immediate environment biased particip- interesting support for this possibility. In ants in some manner to recall disproportion- their study, they measured each participant’s ately more positive memories. This does not, unique capacity to inhibit distracting memo- however, appear to be correct. For example, ries via a simple laboratory version of the Suedfeld and Eich (1995) did a surprising retrieval-­induced forgetting procedure involv- experiment in which they asked participants ing word lists (see Chapter 9 for a description to float quietly in a sensory deprivation tank of this method). They proposed that if people for a full hour, relaxing peacefully; after the tend to inhibit unpleasant memories in daily 366 Autobiographical memory life, then the better someone is at inhibition motivated to limit access to them in autobio- on a laboratory task (i.e., the more retrieval-­ graphical memory (see Skowronski, 2011 for induced forgetting they show), the harder it a discussion of the role of forgetting in should be for that person to come up with self-­enhancement). negative autobiographical memories on an independent test. This is precisely what they found (see Figure 11.7). Interestingly, retrieval-­ induced forgetting did not predict Fading affect bias the recall of positive autobiographical memo- Most people hope to lead a happy life. With ries, which presumably would not have been advancing age, we hope to look back on targets of memory inhibition when they wonderful memories that make us happy and occurred in participants’ lives. have fewer that upset us. The positivity bias But what evidence is there that people are just described suggests one way we make disposed to forget negative memories out of these hopes come true. But what about those emotional self-­ defence? Skoronski, Betz, memories that do survive throughout our Thompson, & Shannon (1991) asked parti- lives? Can we count on happy memories to cipants to record a single distinctive event in retain their capacity to spark joy? What a diary every day over several months. In about negative memories? The psychology addition, those same participants kept a diary literature creates the impression that negative recording events from a friend or relative. At memories are especially powerful, and the end of the experiment, participants were durable in their capacity to make us suffer. tested on their memory for their recorded Indeed, Roy Baumeister proclaimed in the incidents and also for those of their friend. title of an influential paper, the generalization Revealingly, participants showed a significant that in psychology, “Bad is Stronger than positivity bias for their own memories, but Good” (Baumeister, Bratslavsky, Finkenauer, instead showed a significant negativity bias in & Vohs, 2001). Many people intuitively hold memory for their friend’s events, even when this view, envisioning that upsetting events the perceived valence of the events was will inescapably make us upset. matched. This suggests that when unpleasant Fortunately for us, bad is not generally events are self-­ relevant, people are more stronger than good, when it comes to (a) Positive memories (b) Negative memories 1.0 1.0 Autobiographical recall Autobiographical recall 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 r = 0.17 r = –0.31* 0 0 –3 –2 –1 0 1 2 3 –3 –2 –1 0 1 2 3 Retrieval-induced forgetting (z score) Retrieval-induced forgetting (z score) * = statistically significant Figure 11.7 The relationship between retrieval-induced forgetting scores on a laboratory task and participants’ success in retrieving negative or positive autobiographical memoires using the probe technique. People who are better at inhibiting memories generate fewer negative experiences overall (Storm & Jobe, 2012).  Autobiographical memory 367 long-­term memory. This is reflected in a highly counterintuitive and meticulously KEY TERM documented phenomenon known as the Fading affect bias: The consistent tendency for fading affect bias in autobiographical memory. negative memories, over time, to lose affective Put simply, events that make us feel good intensity at a higher rate than positive memories. when they happen to us tend to continue to feel good when they are later recalled, but events that lead us to feel bad when they ones, an effect that grew more pronounced as happen tend to not sting nearly so much more time passed by since the event. when they are later remembered. Thus, This pattern has been replicated exten- across time, negative affect for autobiograph- sively, under a range of conditions. For ical memories fades more rapidly than does example, the fading affect bias occurs regard- positive affect does for positive memories less of whether the event-­related emotions are (Walker & Skowronski, 2009; Walker, active (e.g., elated, angry) or passive (calm, Skowronski, & Thompson, 2003; see Skow- sad). The effect begins to emerge as quickly as ronski, Walker, Henderson, & Bond, 2014 24 hours after an event and can be seen for for an excellent review). In essence, in memories that are decades old. It occurs memory, good can be stronger than bad, in within different ethnicities in the United States that the emotional kiss packed by good mem- (e.g., Caucasion, African-­ American, Native ories tends to outlast the emotional punch American, Latino), with differnet age groups packed by bad memories. (college aged to older adults), and in wide Studies of the fading affect bias very often range of different countries internationally use the diary method (though some have used the probe method). For example, Walker, Vogl, and Thompson (1997) reported three Mean drop in affect intensity studies in which participants kept a diary 2 either for three months (study 1), two years Pleasant (study 2), or nine months (study 3). In all three Unpleasant studies participants recorded a single autobio- 1.5 graphical event on each day and rated it for its emotional valence and intensity on a seven-­ point scale ranging from −3 (extremely unpleasant) to +3 (extremely pleasant), with 0 1 representing neutral. At the end of the diary recording period, the diaries were collected, and the participants returned after an 0.5 extended delay, which ranged from 3.5 months to 4.5 years. Upon return, participants were asked to use the same seven-­point scale to rate how the event made them feel at the 0 time of recall. In each study, they then com- 3.5 months 1 year 4.5 years pared the affect given at the time that the Retention interval event was originally recorded in the diary, and the affect recorded on the final test by sub- tracting the latter from the former. The differ- Figure 11.8 The average reduction in affect rating ence between these two scores provides an between the initial recording of an autobiographical memory in a diary and a delayed assessment of its estimate of how much, in the subjects’ eyes, affective intensity after 3.5 months, 1 year, or 4 years, the emotional response to the event changed separately for positive and negative autobiographical over time. The results of these three studies memories. In general, unpleasant memories are can be seen in Figure 11.8. In every case, the perceived as losing more affective intensity over time intensity of the feeling changed more over time than positive ones (from Walker et al., 1997), which for unpleasant memories than for pleasant illustrates the fading affect bias. 368 Autobiographical memory (e.g., Europe, China, Ghana, New Zealand, according to participants’ prior beliefs (stated and even Maori tribespeople). Most striking of earlier in the day) about how different emo- all, it even occurs regardless of the beliefs of tions fade over time. the participants in the experiment. For This figure plots the difference in affect example, in one study (see Skowronski et al., experienced currently and the experience at 2014), students completed a survey that the time of the original event, separately for assessed their beliefs about memory, including positive and negative events. As can be seen, a question that asked participants about what negative events faded significantly more than they believed happened to emotions in autobi- positive events, which largely retained their ographical memories. The options included: positivity. This fading affect bias arose irre- Positive emotions fade more than negative spective of participants’ beliefs about emotions; Negative emotions fade more than emotion and memory. A similar finding was positive emotions; Positive and negative emo- observed by Ritchie, Skowronski, Hartnett, tions fade equally, and Positive and negative Wells, and Walker (2009), except that parti- emotions get stronger, or Other. Results cipants were asked, when recording memo- showed that 49% of the participants believed ries in a diary, to predict how the event will that positive emotions fade more than negative make them feel in two weeks. Upon returning emotions, whereas 26% believed that positive two weeks later, participants exhibited the and negative emotions fade equally; a mere fading affect bias for negative memories, yet 22% endorsed the view that negative emotions failed to correctly predict the fact that neg- fade more than positive emotions. ative memories would fade more. Interestingly, those same participants Thus, despite our fears about the lasting returned later in the day to complete a retro- damage that negative emotional events may spective memory study in which they recalled cause, we may take some comfort in the fact three positive and three negative events. For that the punch that negative memories pack each, they rated both the affect experienced fades more rapidly over time­—­much more so when the event occurred and the affect they than for the events we feel are pleasant. felt at the time of event recall. The findings of There is truth in the adage that time heals all this test are plotted in Figure 11.9, sorted wounds, it seems. Unfortunately, however, for those suffering from depression or Mean drop in affect intensity anxiety, the fading affect bias appears 2 considerably less strong, and even, in some Pleasant Unpleasant cases, absent (see Skowronski et al., 2014 for 1.5 a review). What leads negative affect to fade more quickly than positive affect, and why might it be deficient in some populations? 1 The answer, at present, is unknown. However, it seems likely to involve the ways in which 0.5 people interact their memories when reminded of them, with positive memories likely to be welcomed and relived, and neg- 0 ative memories suppressed or even con- Negative fades Fades equal Positive fades more more sciously reappraised to help us feel better about them. Such processes are known to be deficient in psychiatric disorders (see, e.g., Figure 11.9 The fading affect bias (greater reduction Engen & Anderson, 2018). Indeed, when we in affect for negative memories over time) occurs, seek meaning in our negative memories, the irrespective of participants’ overall beliefs about how emotional intensity of experiences change over time. negative can at times transmute to positive This suggests that the bias reflects a true change in feelings; indeed, how else could I cheerfully affect for individual memories rather than a simple retell my story (related at the outset of this bias to match one’s ratings to a general belief about chapter) about how I forgot that I had to give memory and affect. Data from Skoronski et al., 2014. a talk about forgetting. Clearly, it would be  Autobiographical memory 369 helpful to examine the mechanisms under- good at estimating the quality of their lying the fading affect bias, to better under- memory, but since they lived locally, he stand what it is that determines whether our agreed to visit. He took along a few standard pasts haunt us or helps us. tests and an autobiographical inventory (AMI), which probes for information from different periods of life. Baddeley was wel- comed by a sociable chap (RM) in his early Variations in forties, who worked in an administrative post Autobiographical in local government. RM’s performance on standard episodic memory tests proved to be Memory Function good but not outstanding, unlike his autobio- graphical memory, which seemed to be rich If you look back over your life, do you feel and quite detailed. However, as he pointed that you have rich memories that you can out, we had no way of knowing whether his recall in detail, highlighting the best and most recollections were accurate. challenging moments of your life? Or, when The next step therefore was to test him you look back, does it feel vague and on verifiable information. One example was general? For example, perhaps you know based on RM’s support for Bradford City, a many facts about your life, and when events local soccer team with a long string of not happened, but it doesn’t feel like you are “re-­ very successful managers. He proved adept at experiencing” whenever you remember recalling managers, together with dates, when things. Perhaps this whole idea of memory as they had been appointed and when sacked, “mental time travel” seems odd to you sometimes coming up with further informa- because it doesn’t feel that way when you do tion as to where he first heard about the it, at least for some of your memories. In fact, sacking. At this point Baddeley decided to there is great variation in the extent, richness, enlist the help of Martin Conway, who is an and level of re-­experiencing associated with expert in that field. Martin drew up a list of autobiographical memories. These variations dates; on half of them some dramatic item of encompass not only ordinary individual news had occurred, and on half this was not differences in autobiographical memory func- the case. Baddeley tested both RM and tions, but also extraordinary cases in which Martin on a sample of such dates. Below is a people have either supernormal or highly typical example: deficient memory, despite being otherwise healthy. Variation in autobiographical AB: “What happened on the 17th memory also can arise from disease or stress. of January 1991?” I describe a range of such cases next, con- sidering the potential implications that this Martin: “Well I have to think to myself variation has for our understanding of auto- what was happening in life at biographical memory. the time. I was a young university lecturer in Lancaster Highly superior autobiographical and we would have been back memory at work after the Christmas break. I taught a course on Over the last decade, remarkable reports have emerged of people with exceptional consciousness. I do remember autobiographical memory capacities. For that (but not much else!).” example, one of the authors, Alan Baddeley, RM: “It was a Thursday and I do was contacted by a lady who claimed that her believe it was the first day of husband had a remarkable memory and won- dered if he would like to test him. Baddeley the Gulf War [correct]. I was was skeptical; people are, in general, not very working for the Department of 370 Autobiographical memory Employment at the time and of notable events. Baddeley checked the date was involved in a training online, and he was correct. When asked about his memory he course at Hebden Bridge, I remarked that for many years he had can’t recall whether I was assumed it was just like everyone else’s. He actually on the training course said that for him dates provided an important or whether I was contributing cue to memories and that this process depended on his capacity to work out the rel- to the training, but I recall the evant day of the week. He reported that his day quite clearly.” memories typically involved visual imagery, AB: “Can you remember hearing mentally observing himself in a particular situation rather than re-­ experiencing the about the news?” event from within. Finally, Baddeley asked RM: “Yes I think I stayed up late to him whether his remarkable memory was watch something on TV and useful in any way. “Not very” he replied, I heard about it before I went “though it makes me a popular member of our pub quiz team!” Probably not much of to bed either later on the an evolutionary advantage for homo sapiens! Wednesday night or the early I have described Baddeley’s testing of hours of Thursday morning.” RM in some detail to give a flavor of the way in which RM remembers. He is very different RM was also good at generating the date from AJ, that I described in Chapter 9. You on which events had happened, provided it may recall that AJ appears to remember every was something within his interest. This day in her life since her teens in intricate included general elections, where he could detail, experienced like a continuous movie provide the date of each British election that is full of emotion. Indeed, AJ describes together with the results in terms of which memories flooding back to her continuously party had attained the majority of seats and during her daily life, as though she was living how large the majority. He could reel off the with a “split screen” with real-­life perception, last 34 UK prime ministers and the last 21 US on the one hand, and memories rolling by, on presidents, mentioning that as a child he had the other. None of