1
Enhancing Cognitive Function in Older Adults
Michelle L. Meade & Denise C. Park
Chapter to appear in: Chodzko-Zajko, W., & Kramer, A. F. (Eds.).
Enhancing Cognitive and Brain Plasticity in Older Adults:
2004 Illinois Physical Activity and Cognitive Functioning Conference.
Corresponding author:
Denise C. Park
Beckman Institute
405 N. Mathews Ave
University of Illinois at Urbana-Champaign
Urbana, IL61801
Telephone: 217-244-1584
Fax: 217-244-9381
Enhancing Cognitive Function in Older Adults
It is well established that aging is accompanied by cognitive decline, and there is currently great interest in understanding contextual moderators of such decline. Improving cognitive function in older adults, even by a small amount, has important consequences for older adults'quality of life, particularly if enhancements defer the age at which susceptible adults become disabled due to Alzheimer’s disease. The current chapter provides a discussion of two approaches aimed at improving cognitive function in older adults. One approach is based on laboratory interventions where subjects are trained in an effort to improve specific cognitive functions, and the other approach focuses on the role that lifestyle variables might play in preserving cognitive function in older adults. After a brief overview of cognitive aging, we discuss both approaches in turn with emphasis given to the overall effectiveness of each approach. Finally, we discuss new research which experimentally manipulates lifestyle variables that we believe offers promising new directions for aging interventions.
Overview of Cognitive Aging
Compelling evidences exists that with age, there are reliable declines in information-processing speed (Salthouse, 1996), working memory capacity (Park et al., 1996; 2002), the ability to task-switch or inhibit irrelevant information (Cepeda et al., 2001; Hasher & Zacks, 1988), and long-term memory function, especially encoding (Craik, 1983; Park et al., 1996; 2002). In addition to the processing declines,there is also evidence that accrued knowledge (which represents the effects of experience) improves with age. The interaction between processing decline and knowledge growth is represented in Figure 1 (Park et al, 2002).
Of interest is the impact that simultaneous processing decline and increased knowledge have on older adults' cognitive function. Recent research suggests that increased knowledge may exert a protective effect on cognition as older adults may rely on this increased knowledge to compensate for processing deficits. For example, Hedden, Lautenschlager, and Park (2005), using structural equation modeling, reported that older adults relied more on verbal knowledge to remember paired associates and produce words in a verbal fluency task while younger adults relied more on speed of processing and working memory capacity.
Along with age-related decreases in behavioral processes, there is also evidence that many neural structures shrink with age. Frontal lobe volume has been shown to decrease across the lifespan, and more moderate shrinkage has been obtained in the medial temporal areas (Raz, 2000). Short-term longitudinal studies have indicated that volumetric changes in brain structure are evident in healthy older adults after only a single year. Specifically, healthy older adults in the Baltimore Longitudinal Study demonstrated brain volume decreases (in both gray and white matter) of an average of 5.4 cubic centimeters per year (Resnick et al., 2003) and also reliable increases in ventricular volume (fluid-filled areas that increase as the brain shrinks) (Resnick et al., 2000).
Despite the decrease in neural tissue with age, it is typically the case that in functional neuroimaging studies, older adults show greater distribution of activation across brain sites than young adults during encoding and working memory tasks and most frequently show increased or bilateral activation in prefrontal areas when young adults show unilateral activation(Cabeza, 2002; Park et al., 2003; Reuter-Lorenz, 2002). There has been considerable theorizing suggesting that this additional activation in the older adults may be compensatory for a declining neural system (Cabeza, 2002; Park et al., 2001;Daselaar et al., 2003; Rosen et al., 2002). In support of this hypothesis, Gutchess et al. (2005) reported that older adults utilized more middle frontal cortex than young to encode pictures they latered remembered, whereas young adults relied more on hippocampal activations. Moreover, there was a direct relationship between increased frontal and decreased hippocampal activations in old but not young subjects. This pattern of findings suggests that older adults compensated for hippocampal processing deficits by recruiting additional resources from the frontal areas.
Evidence from neuroimaging studiesstrongly suggests that older adults may compensate for decline by recruiting different and/or additional neural areas. The flexibility inherent in this compensation suggests that older adults’ neural function is dynamic and that plasticity remains in the neurocognitive system in late adulthood. Moreover, there is a limited behavioral literature which indicates that older adults utilize different and often efficient strategies to compensate for declining cognition (e.g., Hedden et al., 2005; Hertzog, Dunlosky & Robinson, 2005). This reorganization and adaptation is the focus of this chapter. We are interested in understanding ways in which cognitive decline in older adults can be attenuated and devote the remainder of the chapter to intervention strategies for improving cognition in late adulthood. We focus the discussion on intervention programs conducted in the laboratory as well as lifestyle variables and potential interventions that may enhance cognitive function. There is also a large literature on external variables and cues (such as semantic relationships, supportive visual stimuli, list organization) designed to support rather than change cognitive function with age, but we do not consider such stimulus-specific manipulations in this chapter.
Laboratory Based Studies of Active Cognitive Facilitation
Perhaps the most commonly-used laboratory technique designed to enhance cognitive function in older adults is training. In a typical training study, older adults are trained on a specific cognitive task or process in which cognitive decline is evident in order to examine any potential improvement on related tasks. For example, if individuals were trained in a name face memory technique, one might expect improvement in their ability to remember names and faces at a cocktail party, as this task would be very similar to their original training (evidence for near transfer). Of further interest in training studies is whether improvement on the trained task transfers to improvement on novel tasks, such as a name-face training technique improving working memory function or driving (far transfer). The idea here is that the sustained exercise of a particular mental process may strengthen cognitive functionin a variety of domains, much as physical exercise will result in an increase in strength in the specific muscles exercises, as well as general improvements in overall fitness and cardiovascular capacity.
Willis and Schaie (1986) provide evidence that training is effective for improving cognitive function in a specific domain and also stabilizing declining cognitive function. Older adults in this study were classified according to data collected longitudinally over the past 14 years, as either stable or as evidencing decline on spatial orientation and inductive reasoning abilities. All subjects received 5 training sessions of either spatial orientation or inductive reasoning over a 2 week period. Using a construct based approach to measure change, Willis and Schaie (1986) reported significant improvements in both inductive reasoning and spatial orientation as a function of training. More importantly, training stabilized performance for subjects who had previously demonstrated decline on the target abilities and improved performance for subjects whose performance had remained stable. The results suggest that training selectively improved cognitive function, but there was no evidence for general strengthening (note that training did not transfer to novel tasks-- subjects in the spatial orientation condition did not improve in inductive reasoning or vice versa).
In perhaps the most comprehensive training study conducted to date, Ball et al. (2002) enrolled 2800 subjects in training programs for memory function, speed of processing, or reasoning. Subjects were trained over a 6 week period and received 10 one hour training sessions overall and a subset also received additional booster training 11 months later. Subjects were given an immediate performance test as well as yearly follow-up evaluations for two successive years. Of interest was the impact of training on the target abilities (memory, speed of processing and reasoning), but also the impact of training on activities of daily living such as food preparation and financial management (these activities were measured on the Instrumental Activities of Daily Living Scale (IADL), e.g. Willis, Jay, Diehl, & Marsiske, 1992.) Ball et al. (2002) hypothesized that since memory, speed of processing, and reasoning were involved in successful execution of daily living tasks, then improving these component processes should transfer to an improvement in daily living scores. Results indicated that the training improved older adults’ performance on the domain in which they were trained, but there was no transfer to improvement on tasks of everyday living. Moreover, consistent with much prior research on training, Ball et al. (2002) showed benefits of training on the specific ability trained, but no transfer to novel tasks.
Subjects in the Willis and Schaie (1986) and Ball et al. (2002) studieswere classified as healthy older adults. An interesting question concerns the effectiveness of stabilizing cognitive abilities in older adults with memory problems. Cherry and Simmons-D'Gerolamo (2005) provided memory training to older adults with probable Alzheimer's disease. Using a spaced retrieval program in which progressively longer intervals interspersed successful retrieval, the authors demonstrated reliable improvements on immediate tests. Long term effects of spaced retrieval on memory performance of probable Alzheimers' patients is difficult to obtain (Cherry & Simmons-D'Gerolamo, 1999), although those trained previously showed an advantage on re training trials (Cherry & Simmons-D'Gerolamo, 2005). The immediate improvement suggests that training may in some cases impact even cognitively impaired older adults (see Camp, Foss, O'Hanlon, & Stevens, 1996; Camp, Bird, & Cherry, 2000; and Cherry & Smith, 1998 for reviews).
Most recently the impact of training has been examined on neural structures. There are relatively few studies in older adults that have addressed whether cognitive training can result in a permanent change in neural structure or function, although there is evidence that changes can occur in young adults. For example, Draganski et al. (2004) trained young adults on a simple juggling routine over a period of 3 months and found that sustained juggling increased gray matter in the mid temporal area and the left posterior intraparietal sulcus. In a study that involved both young and older adults, Nyberg et al. (2003) demonstrated that mnemonic training increased frontal and occipito-parietal activation in young adults. Older adults also demonstrated neural plasticity, but the pattern was different. Only those older adults whose memory improved with the mnemonic showed increased occipito-parietal activation, and none of the older adults increased frontal activations. Other evidence that there is plasticity with age comes from work by Colcombe et al. (2003) who reported that older subjects who improved aerobic fitness showed more grey matter in the frontal, temporal, and parietal regions compared to sedentary older adults. Although the evidence is relatively sparse at this time, evidence that training permanently increases neural volume or changes neural circuitry for a sustained period will be important in future training work to demonstrate efficacy.
Training provides a relatively effective means of improving cognitive function on the target ability (near transfer) and in some cases has important implications for quality of life (as in memory training for older adults with probable Alzheimer's Disease, or training older adults on techniques that will increase their medication adherence), but important limitations exist in the majority of training studies. As outlined by Marsiske (2005), any improvement from training is relatively short lived, and not all subjects are able to benefit from training. Marsiske also addressed perhaps the biggest challenge facing training studies: demonstrating far transfer. That is, improving or stabilizing a single target ability does not transfer to improvement in other single abilities let alone to a more global cognitive improvement such that older adults might benefit in domains outside of laboratory tasks. Note that one exception is training the Useful Field of View (UFOV) may transfer to better driving ability in older adults (Roenker, Cissell, Ball, Wadley, & Edwards, 2003).
Training studies typically involve sustained practice over a number of days as a mechanism for improving or strengthening cognitive function. Virtually all of the training literature relies on repetition of effortful, top-down cognitive processes that have declined with age with the assumption that repeated performance of the process will make the operation easier of more efficient to perform on a range of tasks. Another approach to improving cognitive function relies on exploiting cognitive processes that are age invariant and do not decline with age (Park, 2000). Such automatic processes (Hasher & Zacks, 1979) utilize few cognitive resources and tend to be bottom-up, data-driven processes that are activated in response to environmental cues or suggestions. Because such automatic processes are particularly sensitive to the environment, they may be useful for enhancing cognitive performance in real-world situations, such as remembering to take medications or implement medical procedures. Liu & Park (2004) relied on activation of automatic processes to improve medical adherence in older adults. They instructed older subjects in the use of a glucose monitor and then indicated that they should monitor their glucose four times a day at specific times for the next three weeks. Subjects were assigned to a condition where they actively rehearsed the intention (engaging an effortful process) or deliberated about pros and cons associated with the action (also effortful). In a third condition, subjects formed "implementation intentions" which involved imagining performing the glucose monitoring behavior being performed in the specific context where they expected to be at the target time on the next day. For example, subjects might imagine themselves drinking orange juice with breakfast and then monitoring their blood glucose after they drank the juice at the appointed time. When subjects actually drank their orange juice the following morning, this act would automatically cue them to also monitor their blood glucose levels. This reliance on automatic processes to enhance adherence yielded clear evidence that the implementation training (which took five minutes) increased use of the glucose monitor at the appointed times relative to the other two conditions. Remarkably, this significant increase was maintained for the entire three weeks of the study. Liu and Park suggested that because older adults led relatively routine lives, once they implemented the behavior, they tended to encounter the same cues in their environment each day which continued to stimulate and maintain the glucose monitoring. These data suggest that another effective technique for improving cognitive function may involve capitalizing on intact automatic processes in older adults as a mechanism for improving function in different situations.
Lifestyle Variables as Predictors of Late Life Cognitive Function
In light of the fact that training older adults on specific cognitive abilities does little to improve overall cognitive function, and that reliance on automatic processes will necessarily be limited to particular situations, it is important to examine other means by which cognition may be enhanced in older adults. One such approach is to examine lifestyle variables. There is some evidence that sustained involvement of various cognitive functions across the lifespan favorably impacts cognitive function in late adulthood. Further, because the activities in real life might be more varied and complex than activities trained in the laboratory, benefit from sustained participation in these activities may be more likely to be supportive of a broad array of cognitive functions.
Research on lifestyle variables assesses cognitive function between groups of people who have been more or less engaged in different types of activities throughout their lives. Of interest is whether those individuals who have been more active in a given activity demonstrate higher cognitive function relative to individuals who have not been active in the same activity. Lifestyle research is largely correlational, leading to interpretive issues (see Salthouse in press, for a discussion), however, the effects are powerful, and across many studies, the correlation exists between leading an engaged life and exhibiting higher cognitive function. Evidence that lifestyle variables impact the rate of cognitive decline in older adults comes from literature examining intellectual engagement in complex work and cognitive activities, engagement in leisure activities, and social engagement. We next review each in turn.
One approach to understanding lifestyle variables on later cognition is to examine the cognitive health of older adults who have spent a lifetime employed in complex jobs relative to individuals who have spent a lifetime employed in less complex jobs. Many years of engaging in the cognitive effort required for complex jobs may exert a protective benefit against later cognitive decline. The evidence concerning whether intellectually challenging environments impact favorably upon cognition during late adulthood is mixed. Many studies support the hypothesis that cognitively demanding work environments foster greater cognitive function in old age. For example, Schooler, Mulatu, and Oates(1999) reported that engagement in “substantively complex” work across the lifespan predicted better intellectual functioning in old age than engagement in less challenging work, even after education and other related factors were controlled. Schooler et al. proposed a reciprocal model that those people who have higher cognitive function are more likely to engage in activities, and the engaged activities foster higher cognitive function, so that there is a feedback loop maintaining higher cognitive function in active older adults. Further evidence showing that complex work may protect against cognitive decline comes from the Maastricht Longitudinal Study. None of the participants enrolled showed cognitive impairment at the beginning of the study, but three years later, 4% of those with low cognitively-demanding jobs showed some cognitive impairment, whereas only 1.5% of those with mentally demanding jobs were impaired (Bosma, van Boxtel, Ponds, Houx, Burdorf, & Jolles, 2003).