2 - Theoretical and Methodological Issues in Memory Training

Editors: Backman, Lars; Hill, Robert D.; Neely, Anna Stigsdotter

Title: Cognitive Rehabilitation in Old Age, 1st Edition

> Table of Contents > Part II - Cognitive Rehabilitation Strategies in Normal Aging > 4 - Multifactorial Memory Training in Normal Aging: In Search of Memory Improvement Beyond the Ordinary

Multifactorial Memory Training in Normal Aging: In Search of Memory Improvement Beyond the Ordinary

Anna Stigsdotter Neely

From everyday functioning ( Where did I put those car keys? ) to the fundamentals of existence ( Who am I?), our ability to remember a previous episode is perhaps one of the most important cognitive assets we have. Throughout human history, possession of a proficient memory has been treasured. The ancient Greeks were devoted to memory improvement, an interest that led to the development of the first memory techniques (Yates, 1966). In addition to the practical advantages of having good memory skills, the fact is that rapid and accurate retrieval of information almost always impresses, whether it is the slick car salesperson who remembers you from an earlier visit or your grandmother who can rattle off the name of nearly every plant and tree in the forest. Good memory can be an important social tool. Thus, it is no wonder that concerns about memory decrements are raised as we get older and begin to experience changes in memory functioning. This concern, in turn, often triggers an interest in what can be done to improve remembering.

In recent years, a fair amount of focus has been on examining how memory can be improved in late life (Verhaeghen, Marcoen, & Goossens, 1992). Some researchers have taken a developmental perspective by addressing questions about age-related differences in terms of memory gain or plasticity (Kliegl, Smith, & Baltes, 1989, 1990; see Verhaeghen in this volume for a detailed review of plasticity research). In the memory-training literature, however, the majority of research conducted on people over 60 has been pragmatic, examining the effects of different memory-training programs (Stigsdotter Neely, 1994; Verhaeghen, 1993). Of particular interest since the late 1970s have been programs that have included training in several memory-relevant skills. The purpose of this chapter is to describe and evaluate these multifactorial

P.64

approaches to memory improvement in old age, including a final section that briefly describes the most common memory techniques and skills that have been used in these programs.

A Multifactorial Approach to Memory Training in Old Age

A common theme in memory-training studies is the acknowledgment that memory improvement, as well as remembering in general, may be determined not only by a limited number of encoding and retrieval processes, but also by other cognitive (attention; Stigsdotter Neely & B ckman, 1993b; Yesavage & Rose, 1983) and non-cognitive factors (self-efficacy beliefs, stress and anxiety levels; Stigsdotter & B ckman, 1989; Lachman, 1991; McEvoy & Moon, 1988; Yesavage, 1985). One of the most influential examples of this contextualistic approach to remembering is the tetrahedral model devised by Jenkins (1979). In short, this model emphasizes that a full understanding of memory performance requires consideration of four basic factors: encoding activities (e.g., organization, repetition, mnemonics), subject characteristics (skills, verbal ability, attention abilities, beliefs, emotional state), retrieval factors (e.g., type of memory test), and the nature of the materials (e.g., verbal, visual). Of particular importance in this model is the notion that these sources of variation interact with each other; delineating the nature of these interactions is thought to optimize the understanding of human memory. In memory-training research, similar models have been advanced emphasizing the multiple origins of memory improvement (see Herrmann & Searleman, 1992, for more details). In this literature, a memory-training program that includes training in several critical areas in addition to training in encoding and retrieval skills is commonly referred to as a multifactorial training program (Herrmann & Searleman, 1992; Stigsdotter & B ckman, 1989).

Many researchers have been enthusiastic about the possibilities of a multifactorial approach to remediating memory loss in old age (Stigsdotter & B ckman, 1989; see West and Lachman, chapters 5 and 6 in this volume). To design a memory-training program addressing several memory-relevant and critical factors is considered more beneficial to memory performance gains in old age than the more dominant approach taken to memory improvement, where only encoding and retrieval skills are trained. In essence, it is hoped that the multifactorial approach can demonstrate performance gains that are stronger, more durable, and generalizable to other tasks and situations than those produced by traditional memory training (Herrmann, Rea, & Andrzejewski, 1988; Herrmann & Searleman, 1992; Stigsdotter & B ckman, 1989). In addition to memory skills, three particular factors that are relevant to memory improvement have caught the attention of researchers interested in the multifactorial view: emotional states (stress and anxiety levels), belief structures (self-efficacy beliefs), and attention skills. A brief presentation follows that outlines these lines of research and explores the effectiveness of memory-training programs in which the multiple origins of memory improvement have been acknowledged. This review will focus on the assumed superiority of multifactorial over traditional training in improving memory performance in old age.

P.65

Investigating the Effectiveness of Pretraining

During the early and mid 1980s, a research team under the direction of Yesavage and Sheikh at the Palo Alto Veterans Administration Medical Center set out to investigate ways to facilitate the learning of mnemonic techniques for older adults. This learning was accomplished by providing pretraining that is, training in a number of different memory-relevant processes before introducing the mnemonic technique. The mnemonic device used most frequently in this work was the face-name mnemonic, which is an imagery-based technique (this and other techniques will be discussed in a later section). The rationale for including pretraining, according to the authors, is exemplified in the following quote from Yesavage, Lapp, and Sheikh (1989): The mnemonic systems commonly used often are confusing, complicated and anxiety-producing for the elderly. It is assumed that preliminary training may prepare them for the task of learning these mnemonics (p. 601). In other words, with mnemonic-relevant pretraining, mnemonic acquisition will be facilitated, and subsequent memory performance should be enhanced. The four pretraining activities investigated were imagery, semantic judgment, relaxation training, and a combination of the two first skills (Hill, Sheikh, & Yesavage, 1988). Both imagery and semantic judgment are critical encoding skills that promote a deep and elaborate processing of the to-be-remembered information (Craik & Lockhart, 1972). Furthermore, these skills are also relevant for augmenting the efficacy of the face-name mnemonic. The ability to form vivid mental images is a crucial aspect of the face-name mnemonic. Making a semantic judgment, which in these studies consisted of producing statements concerning the pleasantness of the face, was believed to increase the elaboration of the to-be-remembered name and face. Relaxation, on the other hand, is thought to enhance attention and reduce high stress or anxiety levels and thereby improve memory (Yesavage, 1984, 1985; Yesavage & Jacob, 1984).

The results of this research demonstrated that pretraining in imagery and semantic judgment improved memory performance more than control group conditions that consisted either of being retested or of hearing lectures on memory and aging designed to promote a positive attitude toward aging an outcome that would be predicted from the extensive literature stemming from the levels of processing tradition (Yesavage, 1983; Yesavage, Rose, & Bower, 1983). What is noteworthy about this research is not the result that pretraining in encoding skills gave rise to improved memory performance following training, but the finding that relaxation training was as effective as imagery and semantic judgment pretraining in facilitating mnemonic acquisition (Hill et al., 1988; Sheikh, Hill, & Yesavage, 1986). Reduced anxiety levels were shown to be especially effective for people who were initially highly anxious (Yesavage, Sheikh, Decker Tanke, & Hill, 1988). These results are in agreement with the notion of a multifactorial view of gain in memory performance in old age, showing that training in cognitive as well as noncognitive skills can have a positive effect on memory gain.

The Relevance of Pretraining to Efficient Mnemonic Learning

From a practical standpoint, it is important to ask whether these results indicate that pretraining is a necessary prerequisite for efficient mnemonic learning. From the

P.66

above results, it is not possible to answer this question with any certainty. They do, however, suggest that the benefits of pretraining are minimal. When interpreting results, one has to consider the context of how the issue at stake was investigated. The positive effects of pretraining are probably restricted to very defined situations. As will be elaborated on further in the next section, pretraining may be especially useful when the mnemonic is complex and the training program is rather brief, as was the case in the above studies, where only 8 hours of training were provided. With more training and cognitively less taxing mnemonics, pretraining may not be as beneficial. Another issue is whether pretraining plus mnemonic training, as conducted in the above studies, actually results in better performance gains than simply receiving the same amount of only mnemonic training, a comparison not made in the above research (the mnemonic group employed in these studies received only 4 hours of training). From a practical point of view, it would be interesting to know how 8 hours of training should be organized to obtain the best improvements on our ability to remember names and faces. For example, is it better to focus exclusively on the complex face-name mnemonic for 8 hours, or is it better to first provide 4 hours of pretraining in imagery, semantic judgment, or relaxation and then to introduce the face-name mnemonic. It is unfortunate that this comparison has not been made because it is a rather critical test of the pretraining notion as a facilitative aid for learning a complex mnemonic. As it stands now, there is no evidence to suggest that 8 hours of task-specific training is less (or for that matter more) effective than a program employing 4 hours of pretraining and 4 hours of mnemonic training; thus, it is hard to make a recommendation one way or the other.

Finally, concerns could be raised about the motivation behind introducing the pretraining notion. As cited above, the motivation for providing pretraining was to facilitate the learning of complex and frustrating mnemonic techniques. An interesting solution to the problem of complexity is to develop strategies to cope with the stress that it produces, as was done in the pretraining approach described above. Another and, considering the target group, perhaps more logical solution would be to try to avoid complex and frustrating techniques all together.

In conclusion, for the purposes of this review, the most interesting and noteworthy aspect of the pretraining studies reviewed here is the introduction of training in non-cognitive factors by addressing emotional states with relaxation training in conjunction with memory training. This work has inspired many other research endeavors, especially the next series of studies to be presented.

Multifactorial Versus Unifactorial Memory Training

Influenced by the work of Yesavage and colleagues, Stigsdotter Neely and B ckman at the Stockholm Gerontology Research Center also conducted a series of studies based on the notion of the multifactorial nature of memory improvement. Like the previously described work, this research examined the effectiveness of a multifactorial memory-training program involving training in encoding skills, attention, and relaxation. The major difference between the two approaches is the role of pretraining as a means of improving mnemonic learning. In this regard, the work by Stigsdotter Neely and B ckman emphasizes to a much greater extent the multifactorial nature of

P.67

improvement by combining several critical abilities into one training program with less or no emphasis on pretraining per se.

The rationale for developing a multifactorial program was based on two lines of reasoning. First, efficient application of the above-mentioned skills is beneficial for learning and memory in general. Having efficient encoding skills and good control over attentional functions in conjunction with optimal levels of stress and anxiety have been shown in several studies to exert a positive effect on memory performance (Eysenck, 1985, 1988; Kausler, 1991). Second, these key processes for efficient remembering have also been shown to be age-sensitive (Salthouse, 1991). Older subjects have been shown to use less optimal encoding skill (see B ckman, M ntyl , & Herlitz, 1990, for a review), to have decrements in attention (see McDowd & Birren, 1990, for a review), and to be more susceptible to stress than younger adults (B ckman & Molander, 1986; Molander & B ckman, 1989). From this perspective, we argued that efforts to alleviate decline in episodic memory performance should be multifactorial; that is, these efforts should include training in several critical cognitive and noncognitive factors if they are to obtain reliable, durable, and generalized gains (see Stigsdotter Neely, 1994, for a review).

To examine the effectiveness of our multifactorial training program, we compared it to a more traditional unifactorial training program, where only encoding skills were practiced and no attention and relaxation training were provided. We also compared our program to a cognitive activation training program comprising training in problem-solving skills (Stigsdotter & B ckman, 1989; Stigsdotter Neely & B ckman, 1993a, 1993b, 1995). Thus, the questions of primary interest addressed in these studies were whether multifactorial training gives rise to immediate and more pronounced gains than unifactorial and cognitive activation training and whether these positive gains generalize over time and tasks.

The results of these studies indicated that both multifactorial and unifactorial training produced improved memory performance immediately after training, whereas those in the cognitive activation training and a control group did not show any improvement between assessments. This pattern of results was maintained 6 months as well as 3.5 years after completion of training. Further, the transfer of gains was limited to tasks very similar to the criterion task. The similar patterns of performance following multifactorial and unifactorial training indicate that attentional and relaxation training may not contribute to an improvement in memory performance above training in specific encoding skills (Stigsdotter Neely, 1994).

Our data indicate rather strongly that training-related gains after this rather brief intervention are task-specific. The following findings provide converging evidence for this view. First, training in only encoding skills results in improved memory performance; the skills trained were especially beneficial in mastering a list recall task. Second, the fact that cognitive activation, attention, and relaxation training did not improve memory performance as measured by list recall of concrete words strongly suggests the specificity of brief memory intervention. And finally, transfer was found only for those tasks most closely related to the criterion task.

As discussed in the previous section, it is important to view results in their appropriate context because conclusions depend on how the issue of interest was examined. First, the above conclusions hold true within the boundaries of how these studies

P.68

evaluated the effectiveness of the multifactorial program. In other words, the findings are contingent on how the training was conducted, the type of memory tests used to measure gains, and so forth. Thus, although attention and relaxation training as conducted in our multifactorial memory training program may not be beneficial in improving memory performance, other forms of attention and relaxation training may still show improved memory performance following multifactorial training.

Second, one might ask why we did not obtain any additional effects from relaxation training like those that Hill et al. (1988) documented. In line with the above contextualization of our findings, the discrepancies in results might be due to methodological differences between the studies. As alluded to earlier, it may be that relaxation training has its primary effects early in training when subjects are unfamiliar with the training context and the task. Hence, the subjects benefit more from reduced levels of stress in shorter programs. In the present case, the multifactorial training program was longer and provided more training in less complex encoding skills than the program used in the Hill et al. (1988) study. This state of affairs might have reduced the stress levels, thus eliminating the potential benefits of relaxation training.

Changing Self-Efficacy Beliefs

The third line of research with a multifactorial approach to memory improvement in old age is studies examining the effectiveness of a program involving training that enhances self-efficacy beliefs and promotes positive and adoptive self-conceptions of memory. This research has been conducted primarily by Lachman and collaborators at Brandeis University (in this volume) and by West and colleagues at the University of Florida (in this volume), but it has lately engaged many researchers (see Floyd & Scogin, 1997, for a review). Since this work is presented in great detail elsewhere in this volume, this chapter will only briefly highlight some of the results that are relevant to the multifactorial approach to training.

In short, the rationale behind this approach is that older adults are not as confident about their memory functioning as the young (Lachman, 1990, 1991). Older adults who believe that there is nothing they can do to prevent memory decline may not have the motivation or desire to apply the new strategies to other problems in other settings (Elliott & Lachman, 1989). Further, there is also ample evidence that memory self-assessments and performance in general are positively related, a finding indicating that people holding negative, self-defeating attitudes also perform more poorly (Bandura, 1984, 1989). Hence, it is suggested that a memory-training program designed to identify and improve negative self-efficacy beliefs and dysfunctional attributional patterns, as well as to improve encoding skills, may improve memory functioning to a greater extent than programs addressing only encoding skills.

Memory-training studies targeting these belief structures have looked at (a) how memory training affects self-efficacy beliefs, (b) how self-efficacy training affects beliefs as well as memory performance, and (c) how a multifactorial approach including training in both encoding skills and self-efficacy beliefs affects memory performance as well as memory beliefs. Results concerning the first two issues have been mixed; some studies have found that memory or self-efficacy training has a positive effect on beliefs and memory performance, while others have not (see West and Lachman in this volume for a more thorough review of these studies).

P.69

Of particular interest from a multifactorial perspective is the last approach, investigating whether an integrated approach that simultaneously targets beliefs and encoding skills results in stronger, more durable, and generalized memory gains than only memory training. In the few studies addressing this issue, multifactorial training programs have yielded memory performance improvements similar to those in programs that trained encoding skills (Lachman, Weaver, Bandura, Elliott, & Lewkowicz, 1992; Weaver, Lachman, & Dick, 1990). Additionally, the multifactorial training program has shown greater beneficial effects on sense of control and perceived ability to improve memory, recall confidence, and attributions than traditional memory training (Lachman et al., 1992; Weaver et al.1990). In only one study did the multifactorial training show greater immediate, as well as maintained, gains than traditional memory training (Caprio-Prevette & Fry, 1996). Again, the majority of the results show no advantage for a multifactorial view in attaining stronger memory gains that are generalized over time, tasks, and situations. Noteworthy is that the multifactorial training led to greater increases in beliefs than did training targeting only beliefs and only memory training, a finding supporting the use of multifactorial training to improve self-efficacy beliefs (see Floyd & Scogin, 1997 for a review; Lachman et al, 1992).

The Multifactorial Approach: Expectations of Gains Beyond the Ordinary?

In this brief summary of research endeavors focusing on the multiple origins of memory improvement in late life, the targeting of self-efficacy beliefs, attention, stress, or anxiety levels in conjunction with memory techniques has predominated. The outcome of this research has revealed that memory improvements following multifactorial training are not distinct from those obtained after traditional memory training with respect to the magnitude of immediate, long-term, and generalized memory improvements. What has become rather evident is that the majority of studies have shown task-specific effects from training (e.g., improvement is seen only in the tasks or abilities trained). Given these less encouraging results of the multifactorial approach, what should be concluded about the viability of this notion as a guiding principle for memory-training programs for older adults? Evidently, to design a training program with the multi notion in order to improve memory beyond the ordinary (e.g., to produce stronger, more durable, and generalized gains) is not justified according to the reviewed literature. Rather, the task-specificity notion might be a more viable option to model training after; the implication is that a multifactorial approach should be used when improvements in several cognitive and noncognitive functions are desired. Hence, if the main goal is to improve older adults' beliefs and attitudes about the aging process and their memory performance on names, then teaching self-efficacy skills in addition to a memory technique for remembering names is justified.

Of course, the limited impact of the multifactorial approach on memory performance may be a function of multiple factors, including (a) the choice of dependent measures for criterion tasks and/or (b) the limited scope of the intervention relative to the difficulty in changing abilities such as emotion (stress and anxiety levels), beliefs, and thought patterns that have become firmly rooted in the individual's self-concept. Before we discard the multifactorial notion, we need more carefully designed studies that include a broader variety of dependent measures of the specific components

P.70

trained and measures of generalizability over time, tasks, and situations. It is actually remarkable that given the optimism concerning memory improvements of a more durable and general nature following multifactorial intervention that more studies have not incorporated tests of long-term maintenance and transfer.

In conclusion, multifactorial training has not given rise to memory improvements over and beyond those resulting from more traditional approaches targeting only memory techniques. It should be pointed out that all of the programs reviewed here, multi- or unifactorial, involved training in encoding and retrieval skills; thus, these skills may be the dominant contributor to improvements in memory performance. The next part of this chapter will primarily review the most common memory techniques used in memory intervention research with older adults.

Memory Strategies: The Common Denominator for Memory Improvements

That memory skills are the common denominator in training programs is, of course, not surprising. Numerous studies have shown that efficient encoding and retrieval processes are of great importance in explicit memory functioning. Efficient or elaborate processing is accomplished when the to-be-remembered information is thought about by semantically or visually integrating the new information with what one already knows. There are many ways in which material can be elaborately processed. For example, this morning I suddenly remembered that I had to call the bank during my lunch hour. This realization came to me during my usual, hectic morning routine of getting myself and my children fed, cleaned, dressed, and ready to meet the new day. I knew that without some sort of elaboration, I was doomed to forget to make the phone call. Unfortunately, a search for a pen and some paper to write myself a reminder would have been as futile at this point in the morning as trying to convince my 2-year-old to keep her cereal either in her bowl or in her mouth. Instead, I chose a mental strategy: I imagined a financial crisis had hit and people were frantically trying to get into the bank to sell their stocks and take out their savings, while I sat idly by and ate my lunch. I hoped that this drastic scene would trigger my memory later at lunch, and it did.

In his excellent book, Searching for Memory: The Brain, the Mind, and the Past (1997), Daniel Schacter wrote:

If we want to improve our chances of remembering an incident or learning a fact, we need to make sure that we carry out elaborative encoding by reflecting on the information and relating it to other things we already know. Laboratory studies have shown that simply intending to remember something is unlikely to be helpful, unless we translate that intention into an effective elaborative encoding. (p. 45)

In my example, the probability of remembering to call the bank increased substantially after my rather sketchy elaboration of this episode. If, instead, I had just told myself without any further elaboration that I had to remember to call the bank at lunch, I would have been left with a more impoverished memory of this episode and most likely a decreased likelihood of later recollection than my elaborated encoding of that same episode. So, in order to foster a rich and long-lived memory of the past, the importance of elaborative encoding cannot be underestimated.

P.71

The vast majority of memory-training studies with older adults have focused on investigating the effectiveness of internal memory strategies as opposed to external strategies such as note-taking skills and the use of diaries and beepers (Intons-Peterson & Newsome, 1992). The most frequently used memory strategies in this literature are the face-name mnemonic and the method of loci (Stigsdotter Neely, 1994; Verhaeghen, 1993). Recently, a fair amount of research has been devoted to the number mnemonic, and several studies have used a variety of more basic memory-encoding skills such as visual imagery and organization. Elaborative encoding is a critical ingredient of nearly all memory improvement techniques. Memory techniques generate rich and detailed encodings that are tightly linked to preexisting knowledge and yet are distinctly different from other items in memory. The remarkable memory feats observed after employing a mnemonic such as the method of loci are certainly due to the technique's capacity to produce elaborate encodings.

In the section that follows, I will give a brief presentation of the most frequently used mnemonics and the more basic memory skills used in this research for remembering single-item information, names, and numbers.

Methods That Improve Memory for Single-Item Information

The method of loci is among the most widely used techniques in the aging literature on memory improvement. It is also one of the oldest techniques known: It was developed by the ancient Greek orators exclusively for memorizing speeches (Yates, 1966). In present time and context, the method of loci has been used to remember specific single-item information in sequence, such as items on a shopping list or things that one has to do during the day (Lorayne & Lucas, 1974). The first step in learning this technique consists of generating a list of distinct places (loci), such as places in one's own apartment or specific locations in town. This sequence of locations is fundamental to this mnemonic and must be overlearned in order for the technique to be useful. To encode information, the first to-be-remembered item is associated with the first locus on the list, the second item to the second locus, and so forth. At retrieval, one returns to the first place in one's mental list of loci, which then serves as a cue for recalling the to-be-remembered item that was placed there. One then goes on to retrieve the second locus, and so forth.

To take a concrete example, let's say that I have to do three things today: call the bank, buy milk, and return a book to the library. My three first loci are on the stove, on the living-room sofa, and in the freezer. I then place the first item I have to remember calling the bank on the stove. Here I picture the turmoil of the stock market taking place on a burning stove, hundreds of people running around screaming, Sell and Buy. The second item to remember is the milk, so I picture a sofa cover with the pattern of cows and pillows designed to look like milk cartons. Finally, in the freezer, I put the books. Later, during the day I consult my loci list. I revisit the stove, the sofa, and the freezer, and hopefully these cues will be efficient enough to take me back to my mental pictures of the things that I needed to get done, namely, banking, buying milk, and returning books.

As mentioned earlier, the first step in learning the method of loci is devoted to generating and memorizing the loci list. The list of loci can either consist of self-generated

P.72

loci or be provided by the teacher; both types of lists have been used in the literature with positive results (Baltes & Kliegl, 1992; Stigsdotter & B ckman, 1989). On one hand, a self-generated list may be preferred over an externally provided list. The reasons are twofold. First, the fact that the list was self-generated is, from a memory perspective, a beneficial and desirable feature since self-generated items are for the most part easier to recall (B ckman & M ntyl , 1988; M ntyl , 1986). Second, the familiarity of self-generated items is an aspect that would increase the likelihood of long-term maintenance of the loci.

On the other hand, there are two aspects worth emphasizing concerning the advantages of a teacher-provided list. First, with a fixed set of loci, it is easier to monitor progress in acquiring the loci list, which is an important feature in a training context. Second, it is a great advantage that the predetermined nature of the loci list lends itself more easily to experimental manipulations, where the association between the loci and the to-be-remembered item can be varied (e.g., Thompson & Kliegl, 1991). The choice of using either a self-generated or a teacher-provided list should be guided by the intention of the training.

There are other less cognitively demanding means of remembering item-specific information, such as executing elaborative encoding on the items or events that one would like to remember, much as is described in the initial example of elaborative processing (calling the bank at lunch). By using visual imagery, a more basic encoding skill, I created a very vivid mental scenario involving a hypothetical, although in these days likely, situation in the bank. Another common method used to remember several items is the link method. In this technique, one associates each to-be-remembered item with its predecessor by having them interact. Imagine that three errands have to be run: mailing a letter, getting a new key made, and dropping some film off for developing. Picture a mailbox with a huge key protruding from the box opening. Then form an association between the key and film developing, such as seeing your pictures being developed portraying only keys, keys, and keys. This technique does not provide the same amount of retrieval support as the method of loci. Here, remembering one item will trigger the memory of the others. These two latter methods can be varied to suit the user and the task by mentally picturing several interacting items in one picture instead of imagining them one by one or two by two (Stigsdotter Neely & B ckman, 1993b). Another common instruction to facilitate recall of item-specific information is to form category organization, like organizing a grocery shopping list in terms of food categories, dishes, and/or menus. This method is used frequently in everyday life.

Finally, the most obvious means of remembering item-specific information is to write the information down on a piece of paper. Efficient note taking should by no means be underestimated as a proficient memory device, as it is perhaps the most prevalent method for remembering single-item information in everyday life used by both young and old. When teaching older adults the techniques described here it is important to emphasize that internal techniques such as the method of loci should not replace external techniques. Rather, they are complementary. Sometimes note taking is to be preferred over internal techniques, but in other situations, the knowledge of elaborative encoding can come in handy, particularly in those situations when one does not have the time to write the to-be-remembered information down (Intons-Peterson & Fournier, 1986).

P.73

In this literature on memory intervention in normal aging, mnemonics such as the method of loci have been seriously questioned as rehabilitative memory aids for older people. This criticism has been based on the premise that mnemonics are difficult to apply and are inappropriate in most everyday situations. Research has also shown that memory experts, like cognitive psychologists themselves, do not use mnemonics to any great extent in everyday life (Park, Smith, & Cavanaugh, 1990). Despite this criticism, there are some important reasons for including training in mnemonics such as the method of loci for elderly people.

One of the advantages of using the method of loci, for example, is the possibility of demonstrating rather impressive performance improvements in a fairly short time, such as recalling words in serial order both forward and backward certainly a skill of little practical value, but the point here is to be able to demonstrate that great gains can be achieved by engaging in elaborate encoding. This demonstration is an encouraging experience for the learner and important for training success. If training can build confidence by showing that improvements can come about when information is processed elaboratively, that is in and of itself as good as any reason for addressing mnemonics in a training context. Further, the method of loci provides a convenient way to talk about the importance of efficient encoding and retrieval processes in explicit remembering. The source of the memory gains is easily explained to the subject by referring to the rich and elaborate type of processing engaged in at encoding and also by focusing on the list of loci serving as a preexisting knowledge structure that provides retrieval support.

Methods That Improve Memory for Names

To be able to remember names is a valued skill for people of all ages. Names are rather difficult to remember partly due to their abstract nature. Many different techniques are used to improve the ability to remember names (Higbee, 1988). These range from rather complex mnemonics to fairly simple recommendations of use-it-or-lose-it. The more complex techniques usually rely on imagery transformations, such as the name-face mnemonic used in the studies by Yesavage and colleagues described earlier (McCarty, 1980; Yesavage, 1983). This name-face mnemonic consists of three steps: First, select a prominent feature of the face; second, make a concrete imagery transformation of the person's name; and third, form an interactive visual image linking the name transformation to the prominent facial feature. This could be done in the following way: I have a colleague with distinctively blue eyes and her name is Pernilla Hegg. To make a mental picture of her name, I take the initial letter from her first name, P and the three last letters of her surname, (H)egg, and combine them into a new word that could be pictured as a gigantic Peg(g). Then I merge Steps 1 and 2 by hanging her distinctively blue eyes on the giant peg. The next time I encounter Pernilla, I will most likely pay attention to her blue eyes, which will then serve as a retrieval cue for my mental representation of her name, blue eyes hanging from a giant peg. Then, I must decode the image to retrieve the name, P for Pernilla and eg for Hegg. This is memory acrobatics at its best! While this method is initially very time-consuming and sometimes difficult, it is usually fun, especially when one creates these images together with other people.

There are many ways in which this complex technique can be simplified without

P.74

jeopardizing improvements. One version consists of only performing the second step in the above sequence. Usually, the problem is to remember the name and not who the name belongs to. Hence, the first as well as the third step may be redundant and could be omitted. The purpose of associating a prominent facial feature with the name is just to guarantee that the name will be given to the right person. So it may be enough for later successful recollection to execute only Step 2, elaborating the name by transforming the name into a visual representation, a concrete picture.

Other less cognitively demanding strategies often used are different strategies for elaborating the to-be-remembered name by associating it with previous knowledge, such as thinking of somebody you know who has that same name. Further elaboration upon this association is also recommended by either reflecting upon the nature of the association or to extend the elaboration to include information that would facilitate later recollection of the information. For example, today when I called my son's dentist, she introduced herself as Gun Persson. Not having a pen around, I thought of my high school friend called Gunnel and another friend who also has Persson as her family name. I imagined the two of them together, dressed as dentists with drills in their hands and scary smiles.

Perhaps the best advice is given by the slogan Use it or lose it. Compliance with the beautiful simplicity of this rather rough statement actually has shown to result in remarkable memory improvements in people with rather severe brain pathologies (Bird & Kinsella, 1995; Camp & McKitrick, 1992). To retrieve the name as soon as you have learned it and to repeat the name at increasingly longer intervals is a technique known as spaced retrieval. This technique is described in great detail by both Camp, Bird, and Cherry and Bird in this volume and is a valuable method that can be easily combined with any of the described memory techniques in this chapter.

The described methods of improving name recall are all effective. However, they do differ in their ease of applicability in everyday life. The first name-face technique described in this section is obviously the most cognitively demanding technique and is not recommended in situations requiring attention to some other source of information than remembering the name. The reason is that for most of us, it comes at the cost of either losing track of the attended information while executing the three steps in the face-name mnemonic or hindering our ability to efficiently perform all three steps. Although this is an excellent method when enough time is available, the less cognitively demanding techniques presented are a better choice in more cognitively challenging situations such as social gatherings.

Methods That Improve Memory for Numbers

Dependency on number codes has increased dramatically in our society since the late 1980s. To gain access to essential services requires remembering personalized number sequences (e.g., four-digit personal identification numbers, door codes). In the memory-training literature, relatively few studies have examined the effectiveness of different number techniques compared to techniques for verbal material. However, recently, a fair amount of work has been conducted in our laboratory investigating, among other things the effectiveness of different methods of remembering numbers (Derwinger, Persson, Hill, B ckman, Stigsdotter Neely, 1999; Hill, Campbell, Foxley, & Lindsay, 1997).

P.75

The number mnemonic used in our research was described by Higbee (1988) and is called the number-consonant or number-phonetic system. Like the method of loci, this is an old technique, which requires a sequence of information to be overlearned for successful recall. The number-consonant technique consists of four steps. The first step requires the memorization of a series of digit-letter pairs, 0 = S or Z; 1 = T or D; 2 = N; 3 = M; 4 = R; 5 = L; 6 = J; 7 = K or G; 8 = F or V; 9 = P or B. It is essential that these digit-letter pairs be overlearned, and several methods of memorization are possible, such as linking a physical characteristic or sound that is common to both the letter and the number. For example, in the case of the number 1, both T and D involve one downward stroke. The second step is the word generation step, which involves transforming the to-be-remembered number string into letters and then joining the letters with placeholder vowels to form a word or series of words. For example, from the number sequence 2194 the letters N T P R are used. A word or word series is then created from this letter combination (e.g., NoT PooR). The third step is to memorize the generated word or word series. Several techniques may be used to facilitate this process, including selecting words that are connected to the number sequence in some meaningful way (e.g., for a bank code with the PIN 0832, the word SaVe MoNey could be created). The fourth step consists of obtaining the original number sequence from the generated word by retracing from Step 3 to Step 1.

In our research, we compared the number-consonant mnemonic with a self-taught program where a group of subjects developed their own techniques for remembering four-digit numbers by using more basic encoding skills, such as visualization and association with previous knowledge. For example, in order to remember the code 4512, the subjects might associate 45 with the end of World War II and 12 with a dozen red roses, creating the mental image of celebrating the end of the war by giving 12 beautiful flowers to a dear friend. Preliminary results of this study showed that both groups improved memory performance after training, but that the number-consonant mnemonic group gained significantly more. This study included a number of transfer tests and a long-term maintenance test. Among the transfer tests that have not yet been analyzed are data pertaining to transfer to everyday life. These data will, of course, be important in evaluating the pros and cons concerning both these methods of remembering numbers.

From a practical perspective, in what situations should this rather complex number-consonant mnemonic be used? From my experience of being a dedicated user and also from reports of a majority of the 120 persons who underwent training with this technique, its greatest use is in remembering critical four-digit codes that are not used every day. Number codes that are used daily quickly become overlearned and proceduralized, which are potent means of long-term remembering, reducing the need for complex elaborative processing.

Aspects to Consider When Designing a Memory-Training Program

When designing and conducting a memory-training program for healthy older adults, there are many things to consider. What behaviors, skills, and processes should be addressed? How much training should be given? What types of tests should be used to measure treatment gains?

P.76

Regarding which components should be targeted in memory training, it was concluded in the first part of this chapter that multifactorial memory training has not yet proven to be a more potent means of obtaining durable and general gains than more traditional approaches focusing exclusively on memory processes. In essence, the majority of research results are task-specific, indicating that the memory improvement is most evident in the task you received training in. This finding suggests that the notion of task specificity might be a viable principle to guide training, meaning that training should specifically focus on improving memory performance on particular tasks and/ or behaviors in particular situations and with particular methods. Thus, if the goal is to improve memory of four-digit numbers, then adequate techniques should be chosen for that purpose, such as the number-consonant mnemonic and/or a less cognitively demanding method such as visual imagery elaboration. However, this does not mean that training need target only one particular task with one particular method. A wider range of tasks, skills, and methods could be used in training, such as memory for names, numbers, and noncognitive factors such as self-efficacy beliefs, but this wider focus has to be accompanied by task-specific training in those particular tasks and skills.

Applying the notion of task specificity when designing a memory-training program also determines the intention behind the intervention, which is to improve memory performance on particular tasks, in particular situations with particular methods, a definition that does not emphasize the generality of results. To have a clearly defined goal as well as the means to obtain that goal is important when introducing the program to the participants. Though our experiences in life tend to support the fact that few skills or improvements come easy, many participants in a memory-training program still have the hope that memory improvement will come easily and lead to general improvements. These positive and unrealistic expectations have to be addressed early in the training program by clearly stating what is going to be accomplished through the training. To adjust beliefs about memory training to the actual intention of training, by stressing that memory training comes at the cost of time and effort and does not lead to general gains, is important for the success of a training program.

Two other issues to consider when designing a memory-training program are the amount of training given and the type of tasks used to measure treatment gains. Concerning how much training is needed, the training phase differs quite extensively in memory-training studies in this area. Usually a training session lasts between 1 and 1.5 hours and the program itself runs anywhere from 1 to 32 sessions, with a mean of around 6 sessions (see Stigsdotter Neely, 1994, for a review). From a practical perspective, six to eight sessions of 60 to 90 minutes each are recommended to adequately master most of the described mnemonics and memory skills. As I pointed out previously, most studies in this area have included only tests of immediate gains in the criterion task and less frequently have used tests measuring transfer over time, tasks, and situations. In order to capture the effectiveness of a memory-training program, all of these measures are important to include.

Finally, the research portrayed in this chapter concerns only healthy aged persons and indicates that most of these people are, cognitively speaking, well functioning compared to subjects with early signs of dementia. To introduce training in cognitively demanding memory techniques for people who experience early signs of dementia or

P.77

forgetfulness beyond the normal is not advisable (see Bird in this volume; B ckman, Josephsson, Herlitz, Stigsdotter, & Viitanen, 1991). A good rule of thumb when designing a memory-training program for practical purposes is to build the training around the interests and cognitive skills of the participants and to calibrate the amount of training accordingly.

Final Conclusions

One of the major goals of applied memory-training research is to understand the complexity of factors behind memory improvement in order to develop efficient memory interventions. As this book attests, memory improvement is a function of several critical factors, such as individual differences, methodological issues, and the skills and processes trained. The aim of this chapter was to focus on the latter of those determinants of memory gain: what skills and processes should be trained to give rise to reliable memory improvements in healthy older adults. I presented research investigating the benefits of a multifactorial approach comprising training in several cognitive and noncognitive factors, compared to memory-training programs where only encoding and retrieval processes have been of primary concern. At this time, the multifactorial approach has not yet given rise to gains beyond the ordinary. As a matter of fact, the critical factor accounting for gains in terms of the skills or processes trained is most likely training of encoding and retrieval operations. This result, in addition to the specificity of gains in other tasks trained, points to the intimate relation between gains and the task trained. To ensure positive benefits from memory interventions, a good piece of advice is to guide training after the notion of task specificity focusing the training specifically on those particular tasks, skills, and behaviors that are of primary interest to the learner.

Reference

B ckman, L., Josephsson, S., Herlitz, A., Stigsdotter, A., & Viitanen, M. (1991). The generaliz-ability of training gains in dementia: Effects of an imagery-based mnemonic on face-name retention duration. Psychology and Aging, 6, 489 492.

B ckman, L., & M ntyl , T. (1988). Effectiveness of self-generated cues in younger and older adults: The role of retention interval. International Journal of Aging and Human Development, 26, 163 167.

B ckman, L., M ntyl , T., & Herlitz, A. (1990). The optimization of episodic remembering in old age. In P. B. Baltes & M. M. Baltes (Eds.), Successful aging: Perspectives from the behavioral sciences (pp. 118 163). New York: Cambridge University Press.

B ckman, L., & Molander, B. (1986). Adult age differences in the ability to cope with situations of high arousal in a precision sport. Psychology and Aging, 2, 133 139.

Baltes, P. B., & Kliegl, R. (1992). Further testing of limits of cognitive plasticity: Negative age differences in a mnemonic skill are robust. Developmental Psychology, 1, 121 125.

Bandura, A. (1984). Recycling misconceptions of perceived self-efficacy. Cognitive Therapy and Research, 8, 231 255.

Bandura, A. (1989). Regulation of cognitive processes through perceived self-efficacy. Developmental Psychology, 25, 729 735.

P.78

Bird, M., & Kinsella, G. (1995). Long-term cued recall of tasks in senile dementia. Psychology and Aging, 11, 45 56.

Camp, C. J., & McKitrick, L. A. (1992). Memory interventions in DAT populations: Methodological and theoretical issues. In R. L. West & J. D. Sinnott (Eds.), Everyday memory and aging: Current research and methodology (pp. 155 172). New York: Springer-Verlag.

Caprio-Prevette, M. D., & Fry, P. S. (1996). Experimental Aging Research, 22, 281 303

Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671 684.

Derwinger, A., Persson, M., Hill, R. D., B ckman, L., & Stigsdotter Neely, A. (1999). Remembering numbers in old age: An investigation of the benefits of mnemonics vs. self-generated memory techniques. Manuscript in preparation. Ume University.

Elliot, E., & Lachman, M. E. (1989). Enhancing memory by modifying control beliefs, attributions, and performance goals in the elderly. In P. S. Fry (Ed.), Psychological perspectives of helplessness and control in the elderly (pp. 339 367). North Holland: Elsevier Science Publishers.

Eysenck, M. W. (1985). Anxiety and cognitive-task performance. Personality and Individual Differences, 6, 579 585.

Eysenck, M. W. (1988). Anxiety and attention. Anxiety Research, 1, 9 15.

Floyd, M., & Scogin, F. (1997). Effects of memory training on the subjective memory functioning and mental health of older adults: A meta-analysis. Psychology and Aging, 12, 150 161.

Herrmann, D. J., Rea, A., & Andrzejewski, S. (1988). The need for a new approach to memory training. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 2, pp. 415 420). Chichester, UK: Wiley.

Herrmann, D. J., & Searleman, A. (1992). Memory improvement and memory theory in historical perspective. In D. J. Herrmann, H. Weingartner, A. Searleman, & C. McEvoy (Eds.), Memory improvement: Implications for memory theory (pp. 8 20). New York: Springer-Verlag.

Higbee, K. L. (1988). Your memory: How it works and how to improve it. New York: Paragon House.

Hill, R. D., Campbell, B. W., Foxley, D., & Lindsay, S. (1997). Effectiveness of the number-consonant mnemonic for retention of numeric material in community-dwelling older adults. Experimental Aging Reasearch, 23, 275 286.

Hill, R. D., Sheikh, J. I., & Yesavage, J. A. (1988). Pretraining enhances mnemonic training in elderly adults. Experimental Aging Research, 14, 207 211.

Intons-Peterson, M. J., & Fournier, J. (1986). External and internal memory aids: When and how often do we use them? Journal of Experimental Psychology: General, 115, 267 280.

Intons-Peterson, M. J., & Newsome, III, G. L. (1992). External memory aids: Effects and effectiveness. In D. J. Herrmann, H. Weingartner, A. Searleman, & C. McEvoy (Eds.), Memory improvement: Implications for memory theory (pp. 101 122). New York: Springer-Verlag.

Jenkins, J. J. (1979). Four points to remember: A tetrahedral model of memory experiments. In J. C. Cermak & F. I. M. Craik (Eds.), Levels of processing in human memory (pp. 429 446). Hillsdale, NJ: Erlbaum.

Kausler, D. H. (1991). Experimental psychology, cognition, and human aging (2nd ed.). New York: Springer-Verlag.

Kliegl, R., Smith, J., & Baltes, P. B. (1989). Testing-the-limits and the study of adult age differences in cognitive plasticity of a mnemonic skill. Developmental Psychology, 25, 247 256.

Kliegl, R., Smith, J., & Baltes, P. B. (1990). On the locus and process of magnification of age differences during mnemonic training. Developmental Psychology, 6, 894 904.

P.79

Lachman, M. E. (1990). When bad things happen to older people: Age differences in attributional style. Psychology and Aging, 5, 607 609.

Lachman, M. E. (1991). Perceived control over memory aging: Developmental and intervention perspectives. Journal of Social Issues, 47, 159 175.

Lachman, M. E., Weaver, S. L., Bandura, M., Elliott, E., & Lewkowicz, C. J. (1992). Improving memory and control beliefs through cognitive restructuring and self generated strategies. Journals of Gerontology, 47, P293-P299.

Lorayne, E. E., & Lucas, J. (1974). The memory book. New York: Stein & Day.

M ntyl , T. (1986). Optimizing cue effectiveness: Recall of 500 hundred and 600 hundred incidentally learned words. Journal of Experimental Psychology: Learning, Memory and Cognition, 12, 66 71.

McCarty, D. L. (1980). Investigation of a visual imagery mnemonic device for acquiring face-name associations. Journal of Experimental Psychology: Human Learning and Memory, 6, 145 155.

McDowd, J. M., & Birren, J. E. (1990). Aging and attentional processes. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (3rd ed., pp. 222 233). San Diego: Academic Press.

McEvoy, C. L. (1992). Memory improvement in context: Implications for the development of memory improvement theory. In D. J. Herrmann, H. Weingartner, A. Searleman, & C. McEvoy (Eds.), Memory improvement: Implications for memory theory (pp. 210 231). New York: Springer-Verlag.

McEvoy, C. L., & Moon, J. R. (1988). Assessment and treatment of everyday memory problems. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory: Current research and issues (Vol. 2, pp. 155 160). Chichester, UK: Wiley.

Molander, B., & B ckman, L. (1989). Age differences in heart rate patterns during concentration in a precision sport: Implications for attentional functioning. Journal of Gerontology, 44, P80-P87.

Park, D. C, Smith, A. D., & Cavanaugh, J. C. (1990). Metamemories of memory researchers. Memory and Cognition, 18, 321 327.

Salthouse, T. A. (1991). Theoretical perspectives on cognitive aging. Hillsdale, NJ: Erlbaum.

Schacter, D. L. (1997). Searching for memory: The brain, the mind, and the past. New York: BasicBooks.

Sheikh, J. I., Hill, R. D., & Yesavage, J. A. (1986). Long-term efficacy of cognitive training for age-associated memory impairment: A six-month follow-up study. Developmental Neuropsychology, 2, 413 421.

Stigsdotter, A., & B ckman, L. (1989). Multifactorial memory training with older adults: How to foster maintenance of improved performance. Gerontology, 35, 260 267.

Stigsdotter Neely, A. (1994) Memory training in late adulthood: Issues of maintenance, transfer, and individual differences. Unpublished doctoral dissertation, Karolinska Institute, Stockholm.

Stigsdotter Neely, A., & B ckman, L. (1993a). Long-term maintenance of gains from memory training in older adults: Two 3 1/2 years follow-up studies. Journal of Gerontology, 48, P233-P237.

Stigsdotter Neely, A., & B ckman, L. (1993b). Maintenance of gains following multifactorial and unifactorial memory training in late adulthood. Educational Gerontology, 19, 105 117.

Stigsdotter Neely, A., & B ckman, L. (1995). Effects of multifactorial memory training in old age: Generalizability across tasks and individuals. Journal of Gerontology: Psychological Sciences, 3, P134-P140.

P.80

Thompson, L. A., & Kliegl, R. (1991). Adult age effects of plausibility on memory: The role of time constrains during encoding. Journal of Experimental Psychology: Learning, Memory, and Cognition, 3, 542 555.

Verhaeghen, P. (1993). Teaching old dogs new memory tricks: Plasticity in episodic memory performance in old age. Unpublished doctoral dissertation, Katholieke University, Leuven, Belgium.

Verhaeghen, P., Marcoen, A., & Goossens, L. (1992). Improving memory performance in the aged through mnemonic training: A meta-analytic study. Psychology and Aging, 7, 242 251.

Weaver, S. L., Lachman, M. E., & Dick, L. (1990). Enhancing memory and self-efficacy: What works and for whom? Unpublished manuscript.

Yates, F. A. (1966). The art of memory. London: Routledge.

Yesavage, J. A. (1983). Imagery pretraining and memory training in the elderly. Gerontology, 29, 271 275.

Yesavage, J. A. (1984). Relaxation and memory training in 39 elderly patients. American Journal of Psychiatry, 141, 778 781.

Yesavage, J. A. (1985). Nonpharmacologic treatments for memory losses with normal aging. American Journal of Psychiatry, 142, 600 605.

Yesavage, J. A., & Jacob, R. (1984). Effects of relaxation and mnemonics on memory, attention and anxiety in the elderly. Experimental Aging Research, 10, 211 214.

Yesavage, J. A., Lapp, D., & Sheikh, J. I. (1989). Mnemonics as modified for use by the elderly. In L. W. Poon, D. C. Rubin, & B. A. Wilson (Eds.), Everyday cognition in adulthood and late life (pp. 598 611). Cambridge: Cambridge University Press.

Yesavage, J. A., & Rose, T. L. (1983). Concentration and mnemonic training in the elderly with memory complaints: A study of combined therapy and order effects. Psychiatry Research, 9, 157 167.

Yesavage, J. A., Rose, T. L., & Bower, G. H. (1983). Interactive imagery and affective judgments improve face-name learning in the elderly. Journal of Gerontology, 29, 197 203.

Yesavage, J. A., Sheikh, J. I., Decker Tanke, E. D., & Hill, R. (1988). Response to memory training and individual differences in verbal intelligence and state anxiety. American Journal of Psychiatry, 145, 636 639.