When most people think about memory, they think about having to remember something over time, whether a few minutes or many days. Everyday life is full of examples—remembering routines, performing on an exam, summarizing a book or movie, and remembering an appointment. These types of situations constitute what memory researchers call long-term memory (Zacks et al., 2000). Long-term memory refers to the ability to remember rather extensive amounts of information from a few seconds to a few hours to decades. Memory researchers have created a wide variety of tasks requiring individuals to remember all sorts of information for varying lengths of time. Over a century of research has indicated that long-term memory represents a relatively large-capacity store in which information can be kept for long periods. More recently, mounting research in cognitive neuroscience suggests that long-term memory is not a unitary construct, but consists of distinct multiple systems (see Chapter 2) that are functionally different and are served by different brain structures.
Let us focus first on the more deliberate and effortful systems of explicit long-term memory. Two important types of long-term memory are semantic and episodic memory. Semantic memory concerns learning and remembering the meaning of
words and concepts that are not tied to specific occurrences of events in time. Examples of semantic memory include knowing the definitions of words in order to complete crossword puzzles, being able to translate this paragraph from English into French, and understanding what the instructor is saying in a lecture. The distinction between episodic and semantic memory is important. Episodic memory is the general class of memory having to do with the conscious recollection of information from a specific event or point in time. Examples of episodic memory include learning the material in this course so that you will be able to reproduce it on an examination in the future, remembering what you did on your summer vacation last year, and memorizing a speech for a play.
Like implicit versus explicit memory, episodic and semantic memory appear to be impacted differently by aging (Ronnlund, Nyberg, Backman, & Nilsson, 2005; Spaniol Madden, & Voss, 2006). For example, episodic memory stays fairly stable until around 55-60 years of age and then shows a precipitous decline beginning around age 65. In contrast, semantic memory increases from 35-55 years of age and then levels off. Although semantic memory starts to decline at age 65, the decline is much less substantial than for episodic memory (Ronnlund et al., 2005).
Semantic Memory. As indicated previously, semantic memory is relatively spared with age in the absence of a disease state. Evidence suggests that there are no deficits in semantic memory processes such as language comprehension, the structure of knowledge, and the activation of general knowledge (Light, 1996; Nyberg et al., 2003). Semantic memory retrieval typically does not tax working memory, and thus older adults can draw upon experience in word meanings and/or general world knowledge. In addition, whereas retrieval of episodic memories is based on cues to the original experience, semantic memories are retrieved conceptually as part of our world knowledge. However, research also shows that although information in semantic memory is retained in older adults, sometimes it is hard to access if it is not exercised on a
regular basis (Hertzog et al., 2003). The fact that semantic memory emphasizes the retrieval of general knowledge is coextensive with a major component of intelligence, acquired knowledge, which is discussed later in Chapter 7. As we shall see, intelligence as acquired knowledge remains stable throughout most of older adulthood.
The major area in which age-related decline in semantic memory can be observed is in its accessibility. This is illustrated in word-finding deficits. Older adults typically have more trouble retrieving a target word when presented with a definition of the word, and they tend to encounter more “tip-of – the-tongue” experiences (Fraas et al., 2002; James & Burke, 2000; MacKay & Abrams, 1996). A tip-of – the-tongue (TOT) experience is when you try to retrieve a name or word and you feel you know the word, but it is not quite accessible. For example, you are at a party and you see someone very familiar; you “know” that person’s name, but you simply cannot retrieve it. Another aspect of this TOT experience is
Older adults have a difficult time remembering even familiar information.
that you can retrieve partial information such as the number of syllables in that person’s name, the initial sounds or letters, and where to stress each syllable (stress pattern). Older adults not only experience more TOT’s, but also report less partial information about the target, both in the laboratory and in everyday life (MacKay & Abrams, 1996). Such TOT problems indicate that even highly familiar information can become more difficult to retrieve as we grow older.
Episodic Memory. Because episodic memory includes so many of the day-to-day activities we adults perform, it has been the focus of more research than any other single topic in memory development (Lovden et al., 2004; Wingfield & Kahana, 2002; Zacks et al., 2000). Typically, researchers study episodic memory by having people learn information, such as a list of words, and then asking them to recall or recognize the items. In a recall test, people are asked to remember information without hints or cues. Everyday examples of recall include telling everything that you can remember about a movie or taking an essay exam. Recognition, on the other hand, involves selecting previously learned information from among several items. Everyday examples of recognition include taking multiple-choice tests and picking out the names of your high school friends from a complete list of your classmates.
Memory researchers use several techniques to study the variables that influence episodic memory performance. For example, they may vary the way that the information to be learned is presented (e. g., in organized groups, with cues, or randomly); the speed at which it is presented; the familiarity of the material; and the conditions for remembering the items (e. g., giving recall cues or making a recognition test easy or hard).
The results from hundreds of studies point to several conclusions. Overall, older adults perform worse than younger adults on tests of episodic memory recall in that they omit more information, include more intrusions, and repeat more previously recalled items (Allen et al., 2002; Lovden et al., 2004; Nyberg et al., 2003; Zacks et al., 2000). These age differences are large; for example, more
than 80% of a sample of adults in their 20s will do better than adults in their 70s (Verhaeghen & Salthouse, 1997). These differences are not reliably lowered either by providing slower presentation or by giving cues or reminders during recall. On recognition tests, older adults do fare better in that differences between older and younger adults are reduced. However, in comparison with young adults, older adults are more likely to accept never- represented items as having occurred on the test especially if they share a conceptual meaning or perceptual resemblance to the previously presented items (Palfai et al., 2003; Verhaeghen et al., 2000; Zacks et al., 2000).
Older adults also tend to be less efficient at spontaneously using internal study strategies, such as using imagery or putting items into categories in one’s mind to organize information during study. When older adults are instructed to use internal organizational strategies such as categorization, however, they not only can do so but also show significant improvement in performance. However, these improvements are not sufficient, in general, to substantially reduce age differences in recall performance (Kahana & Wingfield, 2000; Larsson et al., 2003; Taconnat & Insingrini, 2004; Verhaeghen et al., 1993). These findings suggest that older adults are not as successful in situations requiring them to devise an efficient way to acquire disorganized information, especially when they will be expected to recall it later.
Age differences between older and younger adults can be reduced in several ways. First, allowing older adults to practice or to perform a similar task before learning a new list improves performance. Knowing what one is expected to do usually makes it easier to perform well. Interestingly, better memory performance after practice parallels similar improvements following practice on tests of skills related to fluid intelligence (discussed in Chapter 7). Second, using material that is more familiar to older adults also improves their performance. For example, older adults do not remember words such as byte or Walkman as well as words such as jitterbug or bobbysox. Third, older adults may use compensatory strategies to help themselves remember, which we will examine later in this
chapter. Their beliefs about memory also create an influence on their ability to remember.
What, then, can we conclude about episodic memory? Older adults are apparently disadvantaged when left on their own to face relatively rapid-paced, disorganized information. However, memory performance appears to be somewhat flexible and can be manipulated, with improvements coming from a variety of sources. Later we will consider some attempts to explain why age differences occur and several ways in which memory problems can be corrected, but first we will examine an episodic memory process that is relatively spared with age: autobiographical memory.
Autobiographical Memory. Information that needs to be kept for a very long time (from a few hours to many years) includes facts learned earlier, the meaning of words, past life experiences, and the like. Very little research has been conducted on age differences in such long-term memory for a variety of reasons. For one thing, designing an adequate test of very long-term memory so that we know how to interpret performance is difficult. For example, we often cannot know whether an incident that someone recalls from the past is what actually happened, because we cannot verify the facts. Additionally, if a person does not remember a fact from years past, it may be due either to an inability to retrieve the information or to a failure to have learned the information in the first place. This latter issue is especially important when older and younger adults are compared; we must make certain that both groups had the opportunity and learned the information. Some ingenious researchers, though, have managed to circumvent these problems and have studied a particular aspect of very long-term episodic memory: autobiographical memory.
Autobiographical memory involves remembering information and events from our own life. These recollections provide each of us with a personal history and help define who we are. As important as autobiographical memory is, though, very few studies have looked at how well people remember things over the course of their lives.
Autobiographical memory is primarily a form of episodic memory, although it can also involve
semantic memory. The episodic component of autobiographical memory is the conscious recollection of temporal and spatial events from one’s past. The semantic component consists of knowledge and facts of one’s past (e. g., personal characteristics, knowledge that an event occurred) without having to remember exactly when things occurred and in what order. The episodic details of autobiographical memory are more difficult for older adults, whereas the semantic aspects are much more easily remembered (Barsalou, 1998; Levine, 2004). Accordingly, older adults have been shown to be worse than younger adults in reporting vivid recollections of their past (Levine et al., 2002). However, older adults’ general memories of real-life events have been shown to contain more personal thoughts and feelings and have been rated more interesting than those of younger adults (Hashtroudi et al., 1990; Levine et al., 2002).
In some areas we can assess the accuracy of very long-term memory. Being able to verify what individuals remember accompanied by some record of true events is extremely important in evaluating our ability to remember over long periods of time. Only in cases where records have been kept for many years is this usually possible. Coleman, Casey, and Dwyer (1991) examined records that were available from the Harvard Longitudinal Studies of Child Health and Development on individuals from birth to age 50. Detailed information was collected over the years on such things as which childhood diseases the participants had, whether they smoked cigarettes, and what kinds and how much food they ate. At age 50, participants completed a lengthy questionnaire about these issues, and their responses were compared with similar reports made 10 and 20 years earlier, as well as with the official records. Coleman and colleagues found amazing accuracy for information such as whether a person had ever been a smoker or had a particular disease such as chicken pox. In fact, half of the memories elicited at age 50 were more accurate than the memories for the same information elicited 10 years earlier at age 40! However, information about amounts of food consumed or about individual episodes was not remembered very well. Apparently, these events tend to get blended together and are not stored as separate incidents.
© Mary Kate Denny/PhotoEdit.
Memory for faces from the past can be maintained for long periods of time, as people have experienced at high school reunions as far back as 50 years.
What distinguishes events that are memorable from those that are not? What makes a moment we will remember the rest of our lives? Many people think that highly traumatic events are ones that are indelibly etched in our memories. If so, then people who survived Nazi concentration camps should have vivid memories of their experiences. Wagenaar and Groeneweg (1990) examined the testimony of 78 survivors of Camp Erika, a Nazi concentration camp in The Netherlands during World War II. Dutch police initially interviewed the survivors about their experiences between 1943 and 1948. In 1984, during a war crimes trial for an accused Nazi collaborator, these witnesses gave sworn depositions about their experiences at Camp Erika.
The camp survivors’ recollections were a mix of accurate and inaccurate information. In many cases memory was quite good; even 40 years later about half of the survivors remembered the exact date of their arrival at the camp and their entire identification number. They were able to recall the general conditions of the camp, overall treatment, and the like. However, they also had forgotten many important details, including in some cases their own brutal treatment. Wagenaar and Groeneweg point out that these forgotten details mean that even extreme trauma is no guarantee that an event will be remembered. Perhaps forgetting the horrors of being brutalized is even a type of self-protection.
Events do not always have to be personally traumatic to be highly memorable, though. Some historical events that have considerable personal relevance, very unusual or novel events, and other events that are highly emotional are also remembered very well. Such memories are called flashbulb memories because they are so vivid that it seems as if we have a photograph of the event (Fitzgerald, 1999; Rubin, 1996). Although many events tend to be of major historical significance, such as the assassinations of John F. Kennedy and Martin Luther King, Jr., or the September 11, 2001, terrorist attack, flashbulb memories may also involve personal autobiographical events (Rubin, 1996). Such events tend to impress into memory the circumstances in which the person first heard the news about the event; memories of the circumstances include information about the place, other people present at the time, what activities were occurring, the source of the information, and even associated smells (Cohen, Conway, & Maylor, 1994; Maylor et al., 2002).
An interesting phenomenon arises when you examine vivid autobiographical events across individuals’ life spans. In particular, the age at which a memory is established becomes quite important. As Susan found in the vignette, and as can be seen in Figure 6.1, for both younger and older adults, vivid memories experienced earlier in life (between 10 and 30 years of age) are reported more often than those occurring during middle adulthood (between 30 and 50 years of age; Fitzgerald, 1999; Rubin & Schulkind, 1997; Willander & Larsson, 2006). In addition, on a factual memory test of Academy Award winners, news stories, and teams that played in the World Series, those events that occurred earlier rather than later in life were remembered better (Rubin, Rahhal, & Poon, 1998). Bernsten and Rubin (2002) found that for older adults, the bump in memory was most evident for happiest memories. In contrast, very sad memories showed a decline in recall. It may be that this earlier period of life has importance in defining oneself and thus helps to organize personal memories, but that emotionally negative events are not included as importance is determined (Bernsten & Rubin, 2002; Bluck, 2003; Conway & Holmes, 2004; Pasupathi & Carstensen, 2003).
0 10 20 30 40 50 60 70
Age (years) at time of event
Figure 6.1 The distribution of flashbulb memories produced by younger and older adults.
Source: Reprinted from Social cognitions in aging, T. M. Hess & Blanchard-Fields, Fitzgerald, J. M., Autobiographical Memory and Social Cognition: Development of the Remembered Self in Adulthood, p. 161, Copyright © 1999, with permission from Elsevier.
1. Why is working memory important in understanding age differences in performance?
2. What major differences are there between older and younger adults’ working memory?
3. What are the age-related differences between implicit and explicit memory?
4. What accounts for the fact that implicit memory is spared with aging?
5. How do age differences in episodic and semantic memory compare?
6. What differences have been observed between older and younger adults in aspects of autobiographical memory?