ASSIGNMENT 3
CHAPTER 8
Adult Intelligence: Sketch of a Theory and Applications to Learning and Education
Phillip L. Ackerman
University of Minnesota
OVERVIEW
Intelligence theory and assessment methods have traditionally been aimed at predicting academic success. As such, efforts during the early part of this century first focused on predicting the school success of children and young adolescents (for a review, see Ackerman, 1996). Around World War I, intelligence test content was extended upward—to allow for testing of young and middle-aged adults. As the educational establishment embraced intelligence testing, postsecondary institutions increasingly relied on the use of tests for selection of college and university applicants, starting in the 1920s. Today’s college entrance tests, such as the Scholastic Assessment Tests (SAT) and the American College Testing Program (ACT), show a significant resemblance to the adult intelligence tests of the 1920s. Although these procedures may be useful predictors of college success for young adults, they fail to take account of the differences between child/adolescent intelligence and adult intelligence. A perspective of intelligence that focuses on knowledge as a key ingredient of adult intelligence is presented in this chapter. By moving away from the traditional process-oriented conceptualization of intelligence to a knowledge-oriented conceptualization, many aspects of adult intellectual development can be considered, especially in the context of learning and education for adults. Such a shift in emphasis provides a basis for considering other aspects of the adult learner, such as personality, interests, and motivational skills—and provides a framework for an integrated view of adult development, learning, and education.
In this chapter, I first discuss the differences between child and adult intelligence, as a contrast between process and knowledge components of intellect. Next, a discussion is presented of relations between intelligence and personality, interests, and motivational skills. Putting all of these components together provides for a perspective on adult development that stands in contrast to the traditional view of intellectual decline with increasing age. Finally, some implications of the knowledge-based perspective for adult education and learning are presented.
REVIEW OF DISTINCTION BETWEEN CHILD AND ADULT INTELLIGENCE
Intelligence as Process?
When the first modern procedures were devised for assessing intelligence, Binet and Simon (1905) distinguished between two different approaches, which they called the psychological and pedagogical methods. The psychological method, which they adopted for assessment of children, was specifically oriented toward aspects of intelligence that were believed to be less influenced by cultural privilege—namely memory, reasoning, following directions, and so on. Most of the measures that were developed to assess intelligence were thus process measures. Later developments in individual intelligence assessment have generally adhered to the Binet-Simon formula. Such assessments have been repeatedly shown to be effective instruments for predicting elementary and secondary school success, somewhat effective in predicting postsecondary academic success, and somewhat less effective in predicting occupational success (for reviews, see Ackerman & Humphreys, 1991; Anastasi, 1982).
In contrast, the pedagogical method described by Binet and Simon (1905) attempts to assess intelligence by focusing on what an individual knows—knowledge that is often specific to individuals with differing educational or experiential backgrounds. The pedagogical method essentially focuses on the content of intelligence. A focus on process aspects of intelligence seems logical for prediction of elementary and early secondary academic achievement—when the curriculum is relatively standardized. Ignoring knowledge differences between individuals as they reach adulthood serves only to ignore aspects of intelligence that are increasingly important determinants of postsecondary academic performance, occupational performance, and other attainments in intellectual activities outside of the occupational domain.
In the 1940s, two investigators sought to partially remedy this apparent oversight in mapping intelligence theory to adult development. Hebb (1942) described two different types of intelligence, which he called Intelligence A and Intelligence B. Intelligence A was more of a process- or physiologically based intelligence, and Intelligence B was more cultural/educational/experiential. The preservation of intellectual functioning in adults, even after instances of significant neurological damage, was seen as an expression of the distributed nature of Intelligence B. Similarly, Cattell (1943) divided intelligence into two domains, which he called fluid intelligence and crystallized intelligence. Fluid intelligence (gf) was described as mainly innate and physiological—important for early development and learning, but decreasing as individuals reached adulthood. Fluid abilities include memory and abstract reasoning. Crystallized intelligence (gc), on the other hand, is strictly determined by educational and experiential influences; gc forms out of gf, but unlike gf, gc is maintained well into middle adult ages. The kinds of abilities that are identified as prototypical for gc include verbal comprehension, vocabulary, information, and so on.
Several studies by Cattell and his colleagues (e.g., Cattell, 1963; Horn & Cattell, 1966) have supported the distinction between gf and gc, though not without controversy over the strength of the empirical data on which the demonstrations were based (e.g., see Guilford, 1980; Humphreys, 1967). As far as adult intellect is concerned, though, the main problem with validation of this theory has been with the rather meager sampling that has been done on the diverse domains of knowledge that adults surely have. Cattell (1971/1987) acknowledged this problem—noting that when one attempts to assess adult gc, one either must develop as many tests as there are specialized domains of knowledge, or else one must concentrate on knowledge that has been obtained at adolescence—again, when individuals are assumed to have relatively common educational experiences. To date, researchers have mainly relied on the later strategy—testing vocabulary, reading comprehension, or information that examinees acquire during their adolescent years. Such is the implicit rationale behind the design of general entrance examinations for college and university admissions, such as the SAT or the general portions of the Graduate Records Examination (GRE).
Intelligence as Knowledge
From a pragmatic point of view, prototypical intelligence assessments—that is, those that are predicated on a combination of process measures and high school-type knowledge measures, give adults literally little or no credit for knowledge and skills that they have acquired through occupational or avocational experience. Knowledge obtained through college study or through work experience is simply ignored in such assessments. With this basic feature of intelligence tests in mind, it should provoke little wonder that, on average, adults tend to perform more poorly than high school students on standard intelligence tests. Such results are entirely predicted by the Cattell and Horn theory of Gf and Gc—shown in Fig. 8.1. That is, even though gc shows stable or slightly increasing levels through early and middle adulthood, any gains in gc are more than offset by declines in gf. If performance on intellectual tasks is determined only by the kinds of intelligence that are measured by such measures, we could expect that middle-aged and older adults will substantially underperform their younger counterparts.
However, recent research from the cognitive science and artificial intelligence literature has demonstrated that knowledge is an important determinant of both learning and performance (e.g., Alexander, Kulikowich, & Schulze, 1994). Indeed, the artificial intelligence field, which once supported the notion of a process-oriented “general problem solver” that could handle all kinds of intellectual demands—now has discarded the concept in favor of “expert systems” that are predominantly knowledge-driven (e.g., Schank & Birnbaum, 1994). That is, researchers in these fields have come to understand that what that the individual knows is more likely to determine success or failure in intellectual tasks than either the number of novel facts that can be remembered in a single sitting or the speed of abstract reasoning (although abstract reasoning tests make up a substantial portion of traditional assessments of intelligence—see Anastasi, 1982). Several empirical research programs have provided substantiating evidence on the importance of prior knowledge in content-specific problem-solving situations (e.g., Chi & Ceci, 1987; Chi, Glaser, & Rees, 1982; Glaser, 1991; see Voss, Wiley, & Carretero, 1995, for a recent review). Clearly then, there is a substantial empirical basis for the idea that knowledge is a fundamental component of intelligence (even if one is tempted to adopt a narrow view that only embraces traditional gc abilities—see Cattell, 1971/1987; Horn, 1989).
FIG. 8.1. Hypothetical growth/level of performance curves across the adult life span for traditional measures of gf (fluid intelligence) and gc (crystallized intelligence). Adapted from Horn (1965).
When it comes to understanding the developmental nature of knowledge acquisition and maintenance for adults, the field is too new to provide much hard empirical evidence. However, the few studies in the field (e.g., Holahan & Sears, in association with Cronbach, 1995), tend to indicate that adults often accumulate occupational knowledge well into their middle adult years, and may continue to develop avocational knowledge (such as knowledge about cultural interests and hobbies) well into late adult years. From a theoretical perspective, we can consider the likely developmental trajectories of occupational and avocational knowledge, in comparison to traditional assessments of gf and gc. As shown in Fig. 8.2, quite different images of adult intellectual development will emerge, depending on which aspects of intelligence one chooses to focus on. A focus on knowledge only will likely yield a distinct advantage to adults, whereas a focus on process only will yield an advantage to adolescents and young adults. Any single composite might yield an advantage to either group, depending on the weighting scheme chosen by the investigator. Rather, though, a more comprehensive description of an individual’s intelligence may be derived by developing separate indices for each of these four main components of intelligence.
Unfortunately, no measurement instruments have yet been developed that can gauge where an individual stands on the knowledge components of intelligence. There are many assessment instruments that do assess specific knowledge—such as Advanced Placement exams, College Level Examination Program (CLEP), professional certification and proficiency exams, and the GRE “Advanced” tests. Although such tests are usually administered in isolation (i.e., an examinee typically takes only one or two GRE Advanced tests), these tests are the best current examples of assessment instruments that begin to provide a measure of adult knowledge—at least in the academic and professional domains. Some evidence exists showing that these tests provide a more accurate prediction of postgraduate academic success (Willingham, 1974) and occupational success (Hunter, 1983) than do standard intelligence tests.
The problem for intelligence theorists and practitioners alike, as identified by Cattell (1971/1987), is that to adequately assess knowledge across individuals, one needs to develop many tests, in order to cover the divergent areas of knowledge that adults develop. Although we are still at the early stages of work, one of the major goals of our current research program is to begin to address this shortcoming of adult intellectual assessment. We are starting with academic domains, as measured by the CLEP tests, and supplementing these tests with our own measures of art, music, electronics, technology, and so on. Our goal is not to develop a test for every area of adult knowledge, but rather to broadly sample what it is that people know, and use such measures to study how knowledge develops across the life span. Furthermore, our ultimate hope is to use a knowledge-based approach to intelligence that can better predict intellectual performance of adults than extant measures of intelligence that are predominantly based on assessment of process and high school knowledge.
FIG 8.2. Hypothetical growth/level of performance curves across the adult life span, for intelligence-as-process, traditional measures of gc (crystallized intelligence), occupational knowledge, and avocational knowledge. From Ackerman (1996). Copyright © by Ablex Publishing Corp. Reprinted by permission.
LINKING INTELLIGENCE AND OTHER TRAITS
For children and adolescents, there is relatively little flexibility in the educational curriculum, outside of a small number of elective options (such as music, foreign language, shop, etc.). When it comes to adult learning and education, there is much greater flexibility in both curriculum and external demands on the student’s time. Under these more fluid situations, intelligence, as either process or knowledge, tells only part of the story. The individual’s personality characteristics, interests, and motivational skills also help determine the direction of the individual’s effort, the individual’s likely persistence in a field of study, and along with intelligence, jointly predict the likelihood of success in acquiring new knowledge and skills. Each of these domains is briefly discussed next.
Personality. In a recent review (Ackerman & Heggestad, 1997), we found several personality domains that are related to intelligence. Some personality characteristics appear to be pervasively related to intelligence, both positively (Need for Achievement, Extroversion) and negatively (Neuroticism). However, two related personality traits appear to be especially related to gc and knowledge/achievement, namely a trait called Openness or Culture, and a trait we call Typical Intellectual Engagement (TIE; Goff & Ackerman, 1992). Openness/Culture refers to an individual’s orientation to novel experiences, such as trying new restaurants, going to plays and concerts, and so on. TIE is a term that describes an individual’s orientation toward intellectual activities, such as reading, problem solving, and acquiring knowledge. Neither of these personality traits is much related to process-oriented intellectual abilities (Goff & Ackerman, 1992), but they are positively associated with level of verbal ability and with knowledge about diverse areas of the humanities, arts, and social sciences (Rolfhus & Ackerman, 1996). Assessing Openness and TIE in particular may be expected to help identify adults who are particularly oriented to adult educational experiences.
Interests. In addition to personality traits, individuals differ in their particular interests. Holland (1959, 1973) identified six major areas of interests—Realistic, Investigative, Artistic, Social, Enterprising, and Conventional. Even prior to Holland’s work, interest assessment has been a staple component of matching adults to educational and occupational activities (for a review, see Campbell & Hansen, 1981). The premise of the counseling orientation behind interest assessment is that individuals with similar interests to job incumbents will most likely fit into particular occupations, reflecting job satisfaction (if not actual occupational success). However, a few investigations have found overlap between interests and personality traits (e.g., Lowman, Williams, & Leeman, 1985; Randahl, 1991), and recently we have determined that it is possible to align interests, personality, and abilities—to identify four major “trait complexes,” shown in Fig. 8.3 (Ackerman & Heggestad, 1997). The four trait complexes are Social, Clerical/Conventional, Science/Math, and Intellectual/Cultural. The existence of these trait complexes indicates that, for adults, various intelligence, personality, and interest traits tend to cluster together—in a way that suggests mutually supportive roles among the different domains. Two of these trait complexes (Social and Clerical/Conventional) represent individuals whom we expect to be less likely to pursue adult education and learning opportunities, whereas the other two trait complexes (Science/Math and Intellectual/Cultural) appear to be closely identified with an orientation toward education and learning. However, the Science/Math trait complex is mostly related to process aspects of intelligence—and thus the most challenging for adult—learners, and the Intellectual/Cultural trait complex most closely related to knowledge aspects of intelligence—and thus is probably the most closely identifiable with particularly fruitful domains of adult education.
Motivational Skills. Recent research by Kanfer and her colleagues (e.g., Kanfer & Ackerman, 1996) has identified two major sources of motivational skills that appear to be essential ingredients to learning and skill acquisition, namely Emotion Control and Motivation Control (see also Kuhl, 1985). Emotion Control refers to the skills that individuals use to maintain a focus of attention on difficult and novel tasks, when failure is frequent and frustrations common. A lack of Emotion Control skills leads learners to divert their attention away from the learning task toward emotions, such as worry and evaluation apprehension. In learning situations where individuals have a choice about continuing or abandoning the course of instruction, it can be expected that learners with low Emotion Control skills will be quite likely to express frustration and quit—even as they make progress and improve performance. One of the most common examples of such learning situations has to do with the introduction of new technology—such as VCRs and computer networks.
FIG 8.3. Trait complexes, including abilities, interests, and personality traits showing positive commonalities. Shown are (a) Social, (b) Clerical/Conventional, (c) Science/Math, and (d) Intellectual/Cultural trait complexes. From Ackerman and Heggestad (1997). Copyright © by APA (American Psychological Association).
Although there are methods for ameliorating the demands for Emotion Control (such as scaffolding of instruction and providing a nonthreatening learning environment), probably the biggest positive influences that serve to avoid Emotion Control problems are to present new information in the context of knowledge already available to the learner, or provide explicit information about the nature of expected acquisition trajectories to the learner. In the latter case, it is necessary to take into account the difference between the processing abilities of adults and adolescents, such that when an adult learner is presented with normative information about the kind of progress that is expected from other adults, the learner might obtain a nonthreatening reference anchor (and therefore reduce discrepancies between the expected and actual progress during learning).
Reducing the demands on Emotion Control helps avoid the problems associated with early stages of learning for adults. This is probably the most salient issue in matching instruction to the special capabilities and limitations of adult learners. However, issues of Motivation Control are more subtle and potentially more serious in the long run. Motivation Control refers to an individual’s skills in persisting in a task until mastery is attained—after a rough level of acceptable performance is reached. That is, when learners reach an acceptable threshold of learning or knowledge acquisition, some individuals stop devoting attention to continued learning—which indicates a lack of Motivation Control skills. Learners with Motivation Control skills will maintain effort and attention on a task, even though knowledge acquisition shows diminishing returns over time and effort (Kanfer & Ackerman, 1996). That is, even though the gains in knowledge do not come as quickly to a skilled learner as they do to a novice, the learner with Motivation Control skills has a “mastery” orientation, which will ensure both maintenance of current skills and growth of knowledge over long periods of time. Such skills may even be instrumental in determining which learners become experts or world-class performers in a variety of different intellectual domains (e.g., see Ericsson, Krampe, & Tesch-Romer, 1993).
IMPLICATIONS FOR ADULT DEVELOPMENT
It is possible to provide an integrative perspective of intelligence that takes account of the traditional process components, but also a wider array of knowledge components, along with personality and interest domains. Fig. 8.4 shows one conceptualization, called PPIK—for intelligence-as-process, personality, interest, and intelligence-as-knowledge. This conceptualization combines these four sources of individual-differences variance to yield individual differences in levels of academic and occupational knowledge (Ackerman, 1996). This perspective not only identifies a developmental progression from process to knowledge, but also identifies potential cross-influences between personality and interests and knowledge acquisition. For adults, though, this perspective provides a means for linking traditional measures of intelligence with potential measures of intellectual knowledge and skills. That is, although traditional intelligence measures may partly predict adult knowledge, an adequate assessment of adult intellect requires assessment of adult knowledge. Some areas of knowledge can be adequately measured using existing scales of college-level achievement and occupational proficiency, but such scales only begin to identify adult intellect. Nonetheless, by using a combined assessment strategy that takes account of traditionally measured intelligence, personality, and interests, a more comprehensive evaluation of adult intellect may be possible. Moreover, one can also incorporate aspects of motivational skills into the developmental model, inasmuch as they influence the interface between interests and knowledge acquisition (Kanfer & Ackerman, 1996). Of course, a full understanding of adult development awaits longitudinal evaluation of the changes in knowledge structures that occur across the adult life span.
FIG 8.4. Illustration of the PPIK theory, outlining the influences of intelligence-as-process, personality, interests, and intelligence-as-knowledge during adult development, covering academic and occupational knowledge. From Ackerman (1996). Copyright © 1996 by Ablex Publishing Corp. Reprinted by permission.
IMPLICATIONS FOR ADULT LEARNING AND EDUCATION
There are three specific applications of intellectual assessment for adult educational purposes, namely, selection, classification, and instruction. The PPIK approach adopted here suggests several promising applications across these three fields of application.
Selection. First of all, the PPIK approach to adult intellect suggests that prediction of adult academic success will be improved (i.e., higher validity) when assessments are made of individual differences in relevant knowledge structures, rather than the traditional college entrance examinations. Subject to additional empirical validation (e.g., Willingham, 1974), it is expected that tests of knowledge structures will show higher validity for grades and degree progress, especially as students proceed along the course of knowledge acquisition. Such a result is entirely consistent with the repeated demonstrations showing that, although traditional measures of intelligence well predict first-semester college grades, the validity of these indices declines as students progress through college (Humphreys & Taber, 1973). In contrast, when course material is novel for most students, knowledge tests may have low validity—but as students progress, the knowledge tests are expected to increase in validity. If the ultimate selection criterion is degree completion, higher overall prediction validity may be expected from knowledge tests. In addition, given the developmental progression of knowledge acquisition, middle-aged and older adults may be expected to perform better than younger adults on the knowledge tests—a result that is consistent with the fact that older adults tend to perform better in postsecondary courses than younger adults with equivalent scores on traditional college selection tests, such as the ACT (see, e.g., Sawyer, 1986).
Classification. The task of finding an optimal field of study for adults returning to school is currently more of an art than a science. The counselor will try to integrate work experience information with traditional ability and interest measures. The PPIK approach provides a rationale for finding out, specifically, what it is that the adult learner knows. A profile of knowledge structures (along with an understanding of the knowledge demands of various curricular choices) for prospective adult learners could be used to choose a course of study that optimally builds on the knowledge of each learner. Because adults are likely to have lower levels of process-related intelligence, an assessment along these lines could provide for a scientifically determined “match” between field of study and the learner’s strengths. Moreover, improving the match between adult learners and field of study will help ameliorate problems associated with Emotion Control, by placing the learner in familiar fields of inquiry.
Instruction. Ideal instructional environments match the content and difficulty of instruction to the knowledge and process abilities of the individual learner. When it is impossible to provide one-on-one instruction, it is often possible to provide at least some tailoring of the educational experience to the particular attributes of the learner (e.g., see Snow, 1989). Numerous sources of interactions between individual differences traits and optimal instructional methods have been documented over the past 30 years (e.g., see Cronbach & Snow, 1977; Snow, Corno, & Jackson, 1996; Snow & Yalow, 1982). For most learners, we can expect that taking account of trait complexes (which include intelligence, personality, and interests) can result in more effective instruction. In addition, instructional systems need to take account of the learner’s emotion control skills and motivation control skills, because deficiencies in these skills may influence the likelihood that the learner will persist in a course of study when confronted with inevitable plateaus and failures that accompany any learning situation. Thus, remediation of these motivational skills might preclude actual substantive instruction. However, for adult learners, it may be especially important to take account of age-related decreases in process-related intelligence and increases in knowledge-related intelligence. Particularly appropriate instructional changes would attempt to minimize, for example, rote learning of new facts (which requires process abilities), and maximize the degree to which new material is built on preexisting knowledge structures. Changes to instructional methods might be as simple as increasing the use of analogy examples in the classroom, to exploration of connections between extant knowledge structures and new material. Regardless, the main theme is that with the accompanying changes to the structure of intelligence with adult development, instruction must be adapted away from the current process-based approach toward a knowledge-based approach.
ACKNOWLEDGMENTS
Research reported in this chapter was partially supported by Grant F49620-93-1-0206 from the Air Force Office of Scientific Research, Phillip L. Ackerman, principal investigator. Correspondence concerning this chapter should be addressed to Phillip L. Ackerman, Department of Psychology, University of Minnesota, N218 Elliott Hall, 75 East River Road, Minneapolis, Minnesota 55455 (e-mail: [email protected]).
CHAPTER 9
Mnemonic Strategies for Adult Learners
Russell N. Carney
Southwest Missouri State University
Joel R. Levin
University of Wisconsin
A good memory is a valuable commodity for adult learners, bringing confidence to both social interactions and the workplace. In contrast, difficulty in remembering leads to hesitation and, perhaps, to second thoughts concerning one’s mental state (especially for older adults). Such concerns may even have led to the following exchange:
When the old Indians came in their file to speak to the Governor, he would ask their names; then the governor would ask Ben [Franklin], as he called him, what he must think of to remember them by. He was always answered promptly. At last one Indian came whose name was Tocarededhogan. Such a name! How shall it be remembered? The answer was prompt:—Think of a wheelbarrow—to carry a dead hog on. (Watson, 1830, cited in Pressley & McCormick, 1995, p. 301)
This interesting account illustrates how Franklin recoded the unfamiliar name, Tocarededhogan, in order to make it more concrete, meaningful, visualizable, and hence, more memorable. Techniques such as Franklin’s, which represent “systematic procedures for changing difficult to remember material into more easily remembered material” (Pressley, J. R. Levin, & Delaney, 1982), are referred to as mnemonic strategies, and have been practiced since ancient times (Hrees, 1986; Yates, 1966). Such strategies often facilitate paired-associate learning (e.g., associating names and faces) and serial learning (learning a list of ordered items) in part by the use of interactive imagery (Paivio, 1971). Over the years, a variety of memory-improvement books have recommended mnemonic techniques for learners of all ages (e.g., Bellezza, 1982; Fenaigle, 1813; Furst, 1944; Higbee, 1993; Lorayne, 1990; Lorayne & Lucas, 1974). Additionally, such strategies are routinely presented in self-improvement courses (e.g.,“Dale Carnegie Course in Effective Speaking and Human Relations,” and “Where There’s A Will There’s An A”). Even the ubiquitous Reader’s Digest has offered such strategies to its readership from time to time (e.g., a condensed version of Lorayne, 1985).
In this chapter, we begin by considering the memory concerns of adults, especially the popular notions that (a) information-processing abilities, including memory, decline substantially with age, and (b) there is little that one can do to stop the decline. We then summarize several mnemonic-strategy applications that have, and may be adapted to have, utility for dealing with the memory failures of such learners. We conclude the chapter with some instructional implications stemming from the mnemonic-strategies’ research literature.
MEMORY CONCERNS OF ADULT LEARNERS
As has been observed about the weather, everybody talks about it but no one does anything about it. Likewise, many individuals—particularly older adults—complain about their memories, but do little in an effort to improve them. Such everyday memory complaints are often heightened by the perception that mental abilities decline with age. Indeed, on the basis of an 8-year longitudinal study involving verbal learning, Arenberg and Robertson-Tchabo (1977) concluded that there was some decline after 60 years of age. Yet, Perlmutter and Hall (1992) have observed that although, “… on average, aging is accompanied by a decline in the ability to process information,” this decline is “less severe, later in onset, and true for a smaller proportion of the population than was once believed” (p. 213). This is especially the case for those who continue to be involved in intensive mental activity (Kausler, 1994). Notably, in a large survey of adults over age 55, only 15% said they often had trouble remembering in comparison to 25% who said they never had memory problems (Cutler & Grams, 1988). Likewise, on the basis of their research, Cerella, Rybash, Hoyer, and Commons (1993) argued that aging is associated with minimal decline in cognitive abilities, and that such “counter” positive findings have been underreported.
By the year 2000, it is estimated that 12% of the population will be 65 or older; by the year 2025, 17% (U.S. Senate, Special Committee on Aging, 1987–1988). Despite the positive findings cited earlier, the perception and worry remain among an increasingly gray adult population that normal aging is accompanied by a gradual decline in information-processing abilities, including memory skills. Even worse, perhaps, is these individuals’ perception of the immutability of the process—that is, that there is little if anything that can be done to halt the declining memory parade. True, age-related memory deficits greater than 1 standard deviation below the mean on tests of recent memory have been termed “age-associated memory impairment” (Yesavage, 1990, p. 53), and true, many elderly adults exhibit deficits of that magnitude. But also true, the employment of mnemonic strategies, such as those described in the next section, may enhance memory and, as a consequence, allay the fears (both real and imagined) of adult and elderly learners.
Acknowledging that there is some decline, Perlmutter and Hall (1992) summarized a number of potential explanatory hypotheses. Briefly, the speed hypothesis (Salthouse, 1989) and the generalized slowing hypothesis (Cerella, 1990) both suggest that aging slows down cognitive processing to some extent. A direct consequence of this is that aging adult learners may require more time to process and encode information. The disuse hypothesis (Salthouse, 1989) suggests that as we age, we are less often called upon to use the memory abilities that are tested in the laboratory. Because adults can rely more readily upon external memory aids (Intons-Peterson & Newsome, 1992; Park, Smith, & Cavanaugh, 1990), such as handwritten notes and reminders, they are less likely to make use of associative memory techniques such as those described in this chapter. Simply put, as we age, our memory skills may become rusty through disuse—especially when tested in artificial settings (laboratories) with artificial materials (e.g., lists of unrelated words). The resource reduction hypothesis (Salthouse, 1988) attributes decline in cognitive functioning to a reduction in cognitive resources. Aging may lead to reductions in such resources as mental energy, speed of processing, attentional capacity, and the capacity of consciousness or working memory. As we see later, such reductions, if real, would tend to make the use of mnemonic strategies more difficult—especially if the strategies are complex in nature. Relatedly, it is important to note that although working memory may decrease in capacity, long-term memory seems to remain intact (Poon, 1985). Hence, adults and the elderly have what might be described as a “target-rich” environment for associating new information with prior knowledge. (The “down” side to this is that such a rich environment may in turn contribute to interference during retrieval.) Finally, the inefficient strategies hypothesis (Salthouse, 1988) argues that older adults may tend to select less effective strategies (e.g., rote repetition) for processing information. This explanation is especially appealing, in that memory decline might be offset by training or instruction in more efficient memory strategies. It is to such strategies that we now turn.
ENCODING (OR ASSOCIATIVE) MNEMONICS
A repeated theme in cognitive psychology is that “… learning proceeds most efficiently when to-be-acquired information can be meaningfully related to previously acquired information …” (M. E. Levin & J. R. Levin, 1990, p. 316). In this regard, M. E. Levin and J. R. Levin have proposed that different types of material to be learned can be placed along a “relational processing continuum.” Within this continuum, “efficient” strategies are selected on the basis of the degree of correspondence between new to-be-learned information and the learner’s prior knowledge. In particular, semantic- and schema-based strategies are well suited to the task when the correspondence is high, whereas mnemonic strategies are most beneficial when the correspondence is low. This chapter targets the learning of the latter type of information—information that is mnemonically ripe. J. R. Levin (1983) has proposed that there are three common “R” components of associative mnemonic techniques: recoding, relating, and retrieving. For example, take the task of having to remember that George Washington Carver devoted much of his time to researching the peanut. First, the to-be-associated stimulus name is recoded into something more concrete and familiar (e.g., the name Carver can be recoded as a more concrete, familiar word, such as car). Second, the car and peanuts are related by means of a meaningful, interactive episode. Here, one might imagine a car driving over, and crunching, a bag full of peanuts. Finally, retrieval is accomplished by following the systematic retrieval path that has been established: Carver → car → scene of a car driving over crunching peanuts → peanuts. Levin has termed these steps the “3 Rs” of associative mnemonic techniques. Our primary focus in this chapter is on “encoding” mnemonics (Bellezza, 1981), that is, on strategies that facilitate associative learning. Popular encoding mnemonics (and mnemonic variations) include the face-name mnemonic system, the keyword method, and the phonetic mnemonic system.
The Face-Name Mnemonic Strategy. As was illustrated by the Tocarededhogan example, an everyday task faced by adult and elderly learners is that of remembering people’s names. Although common, name recall represents a difficult task for many individuals, and forgetting them is a frequent complaint of the elderly (e.g., Cohen & Burke, 1993). For example, in a survey of over 100 elderly individuals (Leirer, Morrow, Sheikh, & Pariante, 1990), remembering people’s names was the number one memory skill they wished to improve. In this regard, the face-name mnemonic strategy has been routinely recommended as a useful technique for facilitating memory for people’s names (e.g., Higbee, 1993). The face-name mnemonic involves three steps. Consider, for example, comedian and actor Jim Carrey. The first step is to identify a prominent facial feature, such as his huge grin. The next step is to recode his name, Carrey, into an acoustically similar name clue, such as carry. Finally, an interactive visual image is devised relating the name clue to the prominent feature. For example, one might visualize a pet detective carrying a Cheshire cat with a huge grin. Upon next seeing Mr. Carrey, retrieval proceeds as follows: face → huge grin → interactive image → carrying → Carrey.
The “representational” model of memory for proper names argues that remembering names is difficult because they are both arbitrary and meaningless (Cohen & Burke, 1993). The face-name mnemonic strategy makes the name meaningful by recoding it as a more concrete name clue, and then embedding this clue in a meaningful, interactive image. The ability of interactive visual imagery, in particular, to “glue” items together has been well established, and is theoretically supported by Paivio’s dual-coding hypothesis (e.g., Paivio, 1971). In the end, the procedure yields a systematic retrieval path leading from a pictorial stimulus (the face) to a verbal response (the person’s name).
With few exceptions (e.g., Lewinsohn, Danaher, & Kikel, 1977), research has supported use of the face-name mnemonic technique with undergraduates and adults (e.g., Geiselman, McCloskey, Mossler, & Zielan, 1984; L. D. Groninger, D. H. Groninger, & Stiens, 1995; Hastings, 1982; McCarty, 1980; Morris, Jones, & Hampson, 1978; Patton, 1994; Yesavage & Rose, 1984a). In particular, McCarty’s analysis suggested that all three components of the face-name approach (prominent facial feature, name clue, and interactive image) were essential for the device to be successful. (Note that the Tocarededhogan anecdote does not describe a method of relating the wheelbarrow [“to carry a dead hog on”] to a prominent feature of the individual’s face. Hence, as best we can tell, Franklin was not applying the face-name mnemonic strategy in toto.) Especially relevant to this chapter, Yesavage and Rose investigated the effects of the face-name mnemonic strategy with young (21–38 years old), middle-aged (44–59 years old), and elderly (60–70 years old) adults. They found that the youngest participants remembered the most names, the middle group was intermediate, and the oldest group recalled the fewest names. Nevertheless, all three groups displayed gains in recall after applying the mnemonic strategy. Gruneberg, Sykes, and Hammond (1991), and Gruneberg, Sykes, and Gillett (1994) have successfully used the technique with learning-disabled adults.
Recently, Patton (1994) replicated the positive mnemonic findings with undergraduates when the to-be-remembered stimuli took the form of graduation photographs presented on slides. However, when participants were required to engage in conversation with actual individuals while learning their names, no advantage was gained through use of the face-name mnemonic strategy. Therefore, an important limitation may be that the strategy seems to work only as long as one is able to devote his or her full concentration to the task.1 This appears to be a salient point in that so often our effort to learn names is thwarted by the processing demands involved in greeting someone and making conversation. Nevertheless, there are many instances in which concentrated study is a possibility for adults. For example, an elementary school principal can sit down with a yearbook and study students’ names. Likewise, a minister can peruse a photographic church directory and study parishioner names.
As we have seen, elderly adults often have difficulty remembering people’s names (Cohen & Burke, 1993; Cohen & Faulkner, 1986), and a number of studies have examined the use of the face-name mnemonic strategy in this regard. Jerome Yesavage and his colleagues have been particularly active researchers in this area (e.g., Brooks, Friedman, Gibson, & Yesavage, 1993; Brooks, Friedman, & Yesavage, 1993; Yesavage & Rose, 1984a, 1984b; Yesavage, Rose, & Bower, 1983; Yesavage, Sheikh, Friedman, & Tanke, 1990). This work has generally validated the use of the technique with the elderly, especially in conjunction with what they have termed nonmnemonic pretraining. Such pretraining may focus on relaxation, visual imagery, and semantic elaboration training (Yesavage, 1990).
Although the research points to mnemonic benefits for face-name learning, an adaptation of the strategy may be even more effective when applied to other stimuli that are richer in thematic content than, say, faces. For example, artwork often contains features that can be incorporated into an image involving a recoding of the artist’s name (Carney & J. R. Levin, 1991, 1994; Carney, J. R. Levin, & Morrison, 1988; Franke, J. R. Levin, & Carney, 1991). More generally, it may be possible first to identify an artist’s characteristic style or theme (e.g., Seurat’s pointillism), and then to construct an interactive scene between that and a name clue (e.g., imagining that the painting has been made by dropping tiny drops of syrup [Seurat] all over the canvas). This mnemonic approach may facilitate transfer so that the learner is subsequently able to identify new paintings of similar style or theme (e.g., Seurat’s pointillism), and then to construct an interactive scene between that and a name clue (e.g., imagining that the painting has been made by dropping tiny drops of syrup [Seurat] all over the canvas). This mnemonic approach may facilitate transfer so that the learner is subsequently able to identify new paintings of similar style or theme by the same artist (Carney, J. R. Levin, & Hoyt, 1997). Additional potential applications include labeling outlines of countries in geography, naming parts of the body in anatomy, mineral identification, and identifying unfamiliar animals at the zoo or in the wild (Hoyt, Carney, & J. R. Levin, 1997).
The Keyword Method. The keyword method of vocabulary acquisition (e.g, Atkinson, 1975; Raugh & Atkinson, 1975) is a close cousin of the face-name mnemonic strategy, and is one of the most frequently described mnemonic techniques in educational psychology texts (e.g., Biehler & Snowman, 1997; McCormick & Pressley, 1996; Woolfolk, 1995). Adults engaged in second-language learning or in learning unfamiliar terms related to a new job or interest (e.g., mountaineering) would do well to scrutinize this technique. To illustrate the strategy, consider the geological term bergschrund, which refers to the crack (or crevasse) that forms where the head of a glacier pulls away from the mountain. First, the vocabulary word, bergschrund, can be recoded into a more visualizable, acoustically similar keyword, such as burgers. Next, the keyword and the definition are related by means of a meaningful interactive scene (e.g., an avalanche of hamburgers(bergschrund) tumbling down a mountain and falling into the crack at the head of a glacier). Finally, encoded in this manner, retrieval proceeds as follows: bergschrund → burgers → tumbling hamburgers → crack at the glacier head. More than 20 years of research has demonstrated the usefulness of the keyword method for vocabulary acquisition and related tasks—tasks in which an unfamiliar verbal stimulus prompts a familiar verbal response (J. R. Levin, 1993). A versatile technique, the keyword method has been adapted, extended, and validated as a powerful memory strategy in a variety of situations including: second language vocabulary learning (Atkinson, 1975; Raugh & Atkinson, 1975), acquiring science concepts (J. R. Levin, Morrison, McGivern, Mastropieri, & Scruggs, 1986; M. E. Levin & J. R. Levin, 1990), associating states and their capitals (e.g., J. R. Levin, Shriberg, Miller, McCormick, & B. Levin, 1980), learning about “famous” people (e.g., Shriberg, J. R. Levin, McCormick, & Pressley, 1982) and remembering presidents of the United States (Dretzke & J. R. Levin, 1996), and city attractions (J. R. Levin, Shriberg, & Berry, 1983), to name but a few applications.
Recently, Gruneberg and Pascoe (1996) conducted an experiment with the keyword method, involving a group of healthy older adults whose mean age was about 70. Participants studied 20 Spanish vocabulary words and their meanings. These researchers concluded that “elderly individuals benefit from the keyword method for both receptive and productive foreign vocabulary learning” (p. 108) (although, in the latter instance, the finding was only true given a liberal scoring criterion). Regarding the relevance of the technique for adults, Gruneberg and Pascoe pointed out that “[s]ome individuals may wish to retire to countries where a foreign language is spoken, and these individuals are likely to regard foreign vocabulary acquisition as important” (p. 103).
Wang, Thomas, and their colleagues (Wang & Thomas, 1995; Wang, Thomas, Inzana, & Primicerio, 1993; Wang, Thomas, & Ouellette, 1992) have suggested that individuals using the keyword method experience a faster rate of forgetting (over a delay of several days) than do individuals using a repetition strategy—especially in the absence of an immediate test. However, using a comparable design, Carney, J. R. Levin, Bingham, and Cook (1996) found delayed mnemonic recall advantages after 2- and 5-day delays, even in the absence of an immediate test on the items. At the same time, Carney noted a slightly more rapid decline in the forgetting rate for mnemonically instructed individuals after 5 days. As R. Krinsky and S. G. Krinsky (1996) and others have suggested, the overlearning of mnemonic associations, through additional rehearsal, may be of critical importance in promoting robust long-term mnemonic benefits. It would be interesting to examine this issue with elderly participants.
The Phonetic Mnemonic System. Another common complaint of adult and elderly learners is that it is difficult for them to retrieve numerical information. In the survey mentioned earlier, the second most common memory skill listed as needing improvement by older adults was remembering dates (Leirer et al., 1990). A mnemonic approach recommended by memory improvement books for remembering numerical information such as dates and telephone numbers is the phonetic (or digit consonant) mnemonic system (e.g., Higbee, 1993; Lorayne, 1990; Lorayne & Lucas, 1974). The phonetic mnemonic system is based on a phonetic code whereby numbers are recoded as consonant sounds (e.g., 1 = t, 2 = n, 3 = m, …, 0 = s or z). Because vowels are not used in the code, they may be inserted, as needed, to form familiar words. Thus, the number 20 may be recoded as n + s = nose, 32 as m + n = man, and so forth. Ideally, these words are much more concrete and meaningful—and hence more memorable—than the nominal abstract number (in addition to the critical component of the words then being able to be associated with other information through interactive visual images).
Research evidence regarding the use of the phonetic mnemonic system has been somewhat mixed. Slak (1970) formulated his own memory code, practiced extensively, and then showed improvement in memory span, serial learning, self-paced serial learning, and recognition. Bruce and Clemmons (1982) used the system in a rather complicated procedure for converting between metric and standard measurement units but did not find a mnemonic advantage. Morris and Greer (1984) found a mnemonic advantage in serial recall of a list of two-digit numbers. Patton (1986) found that use of the phonetic mnemonic system actually impaired recall test performance compared to performance by a control group. More recently, Carney and J. R. Levin (1994) combined a simplified version of the technique with the face-name mnemonic strategy to help college students remember “who painted what when”—that is, the dates of various artists’ paintings. They found that memory for dates could be facilitated by recoding the last two digits (all paintings included were from the 19th century) using the phonetic mnemonic system (e.g., with 6 = soft g or j and 1 = t, 1861 could be recoded as jet). The jet, along with a name clue for the artist’s name (e.g., messenger for Meissonier), could then be made to interact with a prominent feature or theme of the painting (e.g., a messenger walking down the steps from his jet in the background). Carney and Levin found a mnemonic advantage for this date-learning technique, in comparison to the learning of students directed to use their own best method.
ORGANIZATIONAL MNEMONICS
Whereas the previous encoding mnemonics facilitate the learning of associated pairs (or clusters) of information, organizational mnemonics (Bellezza, 1981) facilitate the acquisition of ordered information. In particular, older adults are less likely to organize incoming information automatically than are younger adults (Kausler, 1994). Examples of organizational mnemonics include the link mnemonic strategy, the method of loci, and the pegword method. The link mnemonic (Higbee, 1993) is the simplest of these mnemonic systems. Take, for example, the following list of grocery items (found in this order in the store): bread, tuna, milk, corn flakes, and dog food. This mnemonic strategy involves forming sequential interactive images linking each adjacent pair of items (e.g., two pieces of bread wrapped around a tuna; a tuna swimming in a sea of milk; milk being poured over cornflakes; and finally, cornflakes being mixed with dog food. As Higbee pointed out, it is important to make a special effort to remember the first item “starting point” on the list (e.g., linking grocery store to bread in some way). Research has generally supported use of the link mnemonic (e.g., Bugelski, 1977; McCormick & J. R. Levin, 1984; Roediger, 1980).
A second organizational mnemonic is the method of loci. Perhaps the oldest mnemonic method, it dates back to ancient Greece and was used as a memory aid in oratory (Yates, 1966). To apply this strategy, one first selects a number of locations or loci. For example, one might select specific locations encountered sequentially on a walk around campus, such as a specific bench, a statue, a rock wall, and so forth. Next, to-be-remembered items are related to these locations by storing them as interactive visual images. For example, with our ordered grocery list just discussed, one might imagine the bench upholstered with slices of bread, the statue holding a tuna, the rock wall with milk cascading over it, and so on. Finally, retrieval is accomplished by taking a mental walk through these locations, which, in turn, cues the items stored therein. Although the method of loci has been effectively used by college students (e.g., Bower, 1970; Groninger, 1971; Krebs, Snowman, & Smith, 1978), mnemonic benefits with the elderly have been somewhat mixed. For example, Anschutz, Camp, Markley, and Kramer (1985) successfully trained older adults to use the method of loci to remember grocery items. However, a 3-year follow-up to this study found that even though the adults remembered the strategy, they were no longer using it (Anschutz, Camp, Markley, & Kramer, 1987; see Kausler, 1994, for a recent review). Relatedly, a four-seasonal loci approach has been adapted to facilitate students’ learning of the U.S. presidents (J. R. Levin, McCormick, & Dretzke, 1981), and Hwang et al. (1994) used 10 seasonal loci for learning the dates of inventions and the atomic numbers.
A third serial, or ordered recall, mnemonic is the pegword method (e.g, Higbee, 1993). To apply this technique, the learner first memorizes a set of concrete rhyming pegwords, each corresponding to a number from 1 to 10. Thus, 1 = bun, 2 = shoe, 3 = tree, 4 = door, 5 = hive, and so forth, up to 10 = hen. Next, given a list of to-be-remembered items, each of these items is made to interact with the pegword in a meaningful image. Again, let us consider our grocery list. First, one could imagine a package of hamburger buns mashing a loaf of bread in the same section. Next, imagine your shoe kicking a can of tuna. Then, imagine a young tree planted in a plastic milk carton, and so forth. Encoded in this manner, upon entering the grocery store, one simply goes down the list of numbers to cue the desired items. For example, one sounds like bun, and bun brings back the image involving the package of buns mashing the loaf of bread. College students seem to benefit from the technique (e.g., Bugelski, Kidd, & Segmen, 1968), and modifications of the technique have been applied to learning a variety of ordered information, such as presidents (Dretzke & J. R. Levin, 1990), inventions (Hwang et al., 1994), dinosaurs (Mastropieri, Scruggs, & J. R. Levin, 1987), and mineral hardness levels (Scruggs, Mastropieri, J. R. Levin, & Gaffney, 1985). Nevertheless, Kausler (1994), in his review of the literature, described the effectiveness of the technique with elderly learners as “questionable” (p. 106).
Recently, R. Krinsky and S. G. Krinsky (1994, 1996) published studies in which fifth-graders in school settings applied the pegword method to list learning. Their general findings were that although the pegword method produced immediate mnemonic recall advantages over control groups, the mnemonic groups experienced a more rapid rate of forgetting. These findings are in line with those of Wang, Thomas, and their colleagues (Wang & Thomas, 1995; Wang et al., 1992, 1993), which we cited earlier in our discussion of the keyword method. Once again, perhaps overlearning mnemonically acquired information through additional rehearsal is vital for mnemonic benefits to be sustained over the long haul.
INSTRUCTIONAL IMPLICATIONS FOR ADULT LEARNERS
In describing efficient strategy use, Pressley, Borkowski, and Johnson (1987) suggested that “a proficient strategy user knows and can execute a variety of strategies that accomplish many specific cognitive goals” (p. 274). Hence, it would seem to be a simple matter to train adult learners directly to use the well-established mnemonic techniques described in this chapter. One could begin by providing a demonstration of a mnemonic technique to illustrate its efficacy (such as remembering 10 ordered items using the pegword method). Next, various mnemonic techniques could be applied to material pertinent to adults to convince them of the techniques’ relevance (Carney, J. R. Levin, & M. E. Levin, 1994). For example, adults could be asked what is important for them to remember (e.g., medical information), and then the strategies could be tailored to suit their particular needs.
Nevertheless, research has suggested that adults “do not generate mnemonic elaborations reliably in the absence of instruction” (Pressley & McCormick, 1995, p. 301, citing Beuhring & Kee, 1987). Indeed, Park et al. (1990) reported the results of a survey of 69 memory researchers who were asked to rate the frequency of their use of mnemonic techniques. Perhaps surprisingly, the frequency of use was quite low. However, one would suspect that active memory researchers in academe have respectable memories to begin with—and are well stocked with Post-it Notes adjacent to their computers! Additionally, such individuals may be using certain techniques routinely, at some level, without their necessarily being consciously aware of those techniques (e.g., an alphabetic cuing scheme for retrieving a name, leaving a concrete reminder in a strategic place, constructing visual maps, etc.).
Earlier we described the findings of Wang, Thomas, and their colleagues (Wang & Thomas, 1995; Wang et al., 1992, 1993), who have suggested that mnemonically instructed individuals display a faster forgetting rate over a span of several days. Although our research has not supported such a dramatic forgetting rate (e.g., Carney et al., 1996), we have nonetheless noted a slightly faster rate of forgetting for mnemonically instructed students. Again, as R. Krinsky and S. G. Krinsky (1996) and others have speculated, the overlearning of mnemonically acquired associations may be of critical importance in promoting long-term retention. Thus, any training program for adults should provide for additional rehearsal if long-term benefits are to be anticipated. For example, a spaced rehearsal approach (Bjork, 1988; Camp & McKitrick, 1991) might be helpful. To emphasize the importance of this activity in the effective use of mnemonic strategies, we might hereby add a fourth “R” to Levin’s three: recoding, relating, retrieving … and then, rehearsing!
INSTRUCTING OLDER ADULT LEARNERS IN MNEMONIC STRATEGIES
At the beginning of this chapter we listed a number of hypotheses that attempt to explain age-related declines in information-processing abilities, such as memory. Through “intensive and extensive” training in mnemonic techniques (Kausler, 1994, p. 114), we would hope to overcome deficits due to both the inefficient strategies and disuse hypotheses (see also Roberts, 1983). Additionally, it is very important to keep the speed and generalized slowing hypotheses in mind when considering aged learners. Indeed, the “time needed by the elderly to acquire and demonstrate proficiency with a mnemonic technique may need to be extended, particularly if the technique and stimuli are novel to the elderly learner” (Poon, Walsh-Sweeney, & Fozard, 1980, p. 475, cited by Richardson, Cermak, Blackford, & O’Connor, 1987). The slowing of information processing, and the suggestion to provide more practice time is echoed repeatedly throughout the literature (e.g., Finkel & Yesavage, 1989, Pressley & J. R. Levin, 1977; Salthouse, 1985; Treat & Reese, 1976; Yesavage, 1990)—especially regarding more complex mnemonics, such as the method of loci (Yesavage, 1990).
Likewise, the resource reduction hypothesis should be considered. Among other things, the resource reduction hypothesis involves reductions in the capacity of consciousness or working memory. This is problematic, in that short-term memory is “an important determinant of imagery strategy execution” (Pressley et al., 1987, p. 280). As we mentioned earlier, Yesavage (1990) and his colleagues have validated the practice of “pretraining” in teaching the elderly to use mnemonic techniques. Yesavage commented that the pretraining interventions work because “they increase the efficiency of processing” (p. 63), especially for more complex mnemonic techniques. Pretraining consists of three parts: relaxation training, training in visual imagery, and training in semantic elaboration. Relaxation pretraining is analogous to the techniques used in reducing test anxiety. Visual imagery pretraining involves displaying slides, and then having individuals practice visualizing what they have seen. Finally, semantic elaboration pretraining involves asking older learners to make verbal judgments related to their visual images (Yesavage, 1990). These components are designed to offset the finding that the elderly “… often have difficulty applying complex mnemonic strategies because of performance anxiety, difficulty in forming visual images used in associations, and relatively superficial encoding of associated visual images” (Finkel & Yesavage, 1989, p. 199).
Even when older adults are taught to use a strategy, they are less likely to make use of such strategies spontaneously (Camp-Cameron, Markley, & Kramer, 1983), and even when they have discovered benefits in using them, they tend to prefer not to do so (Brigham & Pressley, 1988). As Devolder and Pressley (1992) have demonstrated, young adults are more likely to attribute success to controllable factors, such as strategy use, than are older adults. Additionally, personality traits such as “openness to experience” (Costa & McCrae, 1988) may also play a role in whether an elderly individual learns and successfully uses a mnemonic technique (Gratzinger, Sheikh, Friedman, & Yesavage, 1990). In Costa and McCrae’s model, the open individual is more imaginative than down to earth, prefers variety to routine, and tends to be independent as opposed to conforming (Perlmutter & Hall, 1992).
As illustrated by the preceding points, getting older learners to apply mnemonic strategies is difficult. Taking a pessimistic view, Kausler (1994) suggested that such techniques are “effortful to apply” and “require an imaginal ability that is likely to be difficult for many elderly adults to apply” (p. 114). However, it should be pointed out that the mnemonic benefits we have described are not restricted to visual images per se. For example, in many of our studies with college students, we have routinely provided verbal descriptions of to-be-imagined interactions (e.g., Carney et al., 1988), and have observed comparable mnemonic benefits to those produced by learner-generated images (see also Pressley et al., 1982). Thus, relating to-be-learned items through verbal elaboration (i.e., meaningful sentences tying the items together) might be helpful with learners who seem to have difficulty using visual imagery.
One a more positive note, Kausler (1994) observed that “the keyword method could serve as a means of enhancing the acquisition of a limited foreign language vocabulary that elderly adults could use in visiting a foreign country” (p. 114). More generally, Perlmutter and Hall (1992) have suggested that formal education is no longer reserved for young people. As the proportion of older adults grows larger, and retirement comes sooner, formal education appears to be “spreading across the life span, with middle-aged and older adults enrolling in traditional college programs, in special college programs devised for ‘mature students,’ and in community adult education courses” (p. 417). Our research with mnemonic strategies leads us to be optimistic. We believe that, under the right circumstances, mnemonic strategies can be useful memory techniques for this growing body of graying learners.
