1. Selective Attention

Selective attention refers to the ability to attend to some stimuli while disregarding others that are irrelevant to the task at hand. For example, in visual search tasks, people are asked to search a visual display for a target letter that is surrounded by other nontarget letters. The task can be made more difficult by increasing the similarity of targets and distractors (e.g., search for an F in a background of Es), or by increasing the number of relevant or irrelevant features that are part of the search criteria. In another task — the Stroop task — people are asked to name the color of ink in which an incongruent color word is printed, (e.g., the word “red” printed in green). Here, the word information tends to interfere with color naming, causing errors and an increase in response times. To perform well in these kinds of tasks, people have to select the relevant stimulus or dimensions for processing and ignore the irrelevant ones. Although findings are not entirely consistent across studies and may differ across tasks, in general older adults appear to be slower than younger adults in responding to the targets, but are not differentially affected by distraction [3, 4]. Thus, deficits found in many of these tasks can be largely attributed to a general slowing of information processing in older adults rather than to selective attention deficits per se.

2. Divided Attention and Attention Switching

Divided attention has usually been associated with significant age-related declines in performance, particularly when tasks are complex. Divided attention tasks require the processing of two or more sources of information or the performance of two or more tasks at the same time. For example, people may have to monitor stimuli at two different spatial locations, or they may be asked to make semantic judgments about visually presented words while simultaneously monitoring for the occurrence of an auditorily presented digit [5]. The cost of dividing attention is assessed by comparing performance under dual task conditions to performance when the tasks are performed separately. Results suggest that older adults are more affected by the division of attention than young adults, particularly when the attentional demands of the two tasks are high. In addition, older adults seem less able to allocate resources appropriately when instructions are given to vary task priority [6]. These findings cannot be completely accounted for by a general slowing of information processing, but instead are usually explained in terms of declining processing resources associated with normal aging. Such limited resources are over-extended in older adults when attention must be divided between two or more sources. Similarly, the performance of older adults is slowed to a greater degree than that of young adults when attention must be switched from one task to another, requiring a change of mental set [4].

There is evidence that age deficits in divided attention and attention switching can be reduced by practice or extended training [7] and by aerobic exercise [8]. The exact mechanism of such improvements, however, is unclear. In the case of task-specific training, it is possible that some aspects of the tasks become automatic with practice, thus requiring fewer attentional resources. Alternatively, participants may develop strategies with extensive training that reduce the attentional demands of the tasks. It has been hypothesized that cardiovascular fitness may improve the efficiency of neural processes or may provide increased metabolic resources for task performance. Interestingly, the enhancement effects of aerobic exercise appear to be greatest on tasks involving executive control of attention [9], which depend largely on prefrontal cortex.

3. Sustained Attention

Sustained attention refers to the ability to maintain concentration on a task over an extended period of time. Typically, vigilance tasks are used to measure sustained attention, in which people must monitor the environment for a relatively infrequent signal, such as a blip on a radar screen. In general, older adults are not impaired on vigilance tasks.

4. Attention: Summary and Implications

Older adults show significant impairments on attentional tasks that require dividing or switching of attention among multiple inputs or tasks. They show relative preservation of performance on tasks that require selection of relevant stimuli; and although they are slower than young adults, they are not differentially impaired by distraction. They also are able to maintain concentration for an extended period of time. The tasks on which older adults show impairments tend to be those that require flexible control of attention, a cognitive function associated with the frontal lobes. Importantly, these types of tasks appear to be amenable to training and show benefits of cardiovascular fitness.

Attentional deficits can have a significant impact on an older person’s ability to function adequately and independently in everyday life. One important aspect of daily functioning affected by attentional problems is driving, an activity that, for many older people, is essential to independence. Driving requires a constant switching of attention in response to environmental contingencies. Attention must be divided between driving, monitoring the environment, and sorting out relevant from irrelevant stimuli in a cluttered visual array. Research has shown that divided attention impairments are significantly associated with increased automobile accidents in older adults [3, 10]. Given the previously noted findings of the effects of practice, extended training on driving simulators under divided attention conditions may be an important remedial activity for older people.

1. Theories of Working Memory

Three theories of cognitive aging have been articulated within the context of working memory deficits, although they may apply more broadly across other cognitive domains: (1) one theory proposes a reduction of attentional resources, (2) one focuses on reduced speed of information processing, and (3) one ascribes problems to a failure of inhibitory control (for review, see [16]).

2. Working Memory: Summary and Implications

Older adults exhibit significant deficits in tasks that involve active manipulation, reorganization, or integration of the contents of working memory. Although the mechanisms underlying these age-related deficits are as yet poorly understood, the effects of such deficits are very likely far-reaching. Many complex everyday tasks such as decision-making, problem-solving, and the planning of goal-directed behaviors require the integration and reorganization of information from a variety of sources. It seems likely that attention, speed of information processing, and the ability to inhibit irrelevant information are all important functions for effective performance of these higher-level cognitive tasks. Whether these functions might be subsumed under a domain-general executive controller that is impaired by normal aging —something akin to the central executive in Baddeley’s model of working memory — or whether there may be multiple control processes that are independently affected by aging, is currently an issue under investigation. The brain regions that are active during working memory tasks are also beginning to be identified in a variety of functional neuroimaging studies. Results suggest that different areas are activated in young and old adults, particularly within the prefrontal cortex, indicating that younger and older adults are performing these tasks differently [12]. An understanding of age-related neurophysiological changes may help to account for these differences.

1. Episodic Memory

Episodic memory refers to memory for personally experienced events that occurred in a particular place and at a particular time. This kind of memory allows one to think back through subjective time — what Tulving calls mental time travel [28] —and it usually evokes an “I remember” response. Episodic memory may be distinctly human; it is the most advanced form of memory and is ontogenetically the latest to develop. It also seems the most susceptible to brain damage and the most affected by normal aging.

The episodic memory problems experienced by older adults may involve deficient encoding, storage, or retrieval processes. At the input stage, older adults may encode new information less meaningfully or with less elaboration, so that memory traces are less distinctive, more similar to others in the memory system, and thereby more difficult to retrieve [29]. Alternatively, older people may attend to focal or salient information but fail to take account of peripheral detail, or they may fail to integrate contextual aspects of an experience with central content — what is sometimes referred to as a source memory problem [30]. Many of the common everyday memory lapses reported by normal older adults, such as forgetting where they parked their cars, likely involve poor encoding. These kinds of memory failures have generally been attributed to reduced use of effortful encoding strategies, which depend particularly on prefrontal brain regions. Another possibility is that noticing and integrating the various aspects of an experience involve divided attention and require working memory.

Older adults may also experience problems at the level of storage or consolidation. This aspect of episodic memory critically depends on medial temporal lobe structures, particularly the hippocampus. Consolidation is thought to involve the binding of various aspects of experience into a composite memory trace. What may be particularly critical for episodic memory and impaired in older adults is the extent to which an event is bound to its spatial and temporal context.

Finally, considerable evidence points to retrieval as a source of episodic memory problems in aging. Although it is clear that retrieval is at least partly dependent on encoding (i.e., well-encoded information is easier to retrieve), there are also effortful retrieval processes that appear to be impaired by aging. Older adults tend to show deficits on tests of free recall, to a somewhat lesser degree in cued recall, but minimally in recognition memory. Craik [18] has argued that the requirement to self-initiate strategic search processes in recall taxes the limited resources of older people. To the extent that environmental support can be provided at retrieval as well as at encoding (by providing good cues or using recognition tests, for example), the resource demands of encoding and retrieval are reduced and age differences are minimal. Similarly, Jennings and Jacoby [31] have demonstrated that recollection, which requires effortful retrieval of episodic detail, is impaired with age, whereas the more automatic judgments of familiarity are intact. Evidence from functional neuroimaging and neuropsychological studies suggests that these more strategic retrieval processes depend on the prefrontal cortex, as well as the hippocampus [32, 33].

2. Semantic Memory

Semantic memory refers to one’s store of general knowledge about the world, including factual information such as “George Washington was the first president of the United States” and knowledge of words and concepts. Such information is not tied to the space or time of learning, and its retrieval is generally prefaced with “I know.” Normally aging older adults do not have significant impairments in semantic memory. In fact, their knowledge of the world often exceeds that of young people. In addition, although access to information may be somewhat slower (particularly for words and names), the organization of the knowledge system seems unchanged with age (for review, see [34]). Semantic memories are believed to be stored in a variety of regions in posterior neocortex.

3. Autobiographical Memory

Autobiographical memory involves memory for one’s personal past and includes memories that are both episodic and semantic in nature. The bulk of the evidence suggests that recent memories are easiest to retrieve, those from early childhood are most difficult to retrieve, and there is a monotonic decrease in retention from the present to the most remote past, with one exception. Events that occurred between the ages of 15 and 25 are recalled at a higher rate — what is referred to as the reminiscence bump — a finding that has usually been attributed to the greater salience or emotionality of the memories during this time period. This general pattern holds across all ages, suggesting that autobiographical memory is largely preserved with age (for review, see [35]). More detailed analyses of the nature of the autobiographical information retrieved, however, has suggested that although memory for personal semantics is intact in old age, memory for specific episodic or contextual details about one’s personal past may be impaired. In a recent study, Levine et al. [36] observed that although older adults reported the gist of autobiographical event memories as well as young people, they reported fewer details. There may be exceptions to this finding, however. Recent studies of flashbulb memory have demonstrated that older adults remember as much as young adults about the details and circumstances surrounding highly emotional public events such as the death of Princess Diana or the 9/11 attack on the World Trade Center in New York City [37, 38].

4. Procedural Memory

Procedural memory refers to knowledge of skills and procedures such as riding a bicycle, playing the piano, or reading a book. These highly skilled activities are acquired more slowly than episodic memories through extensive practice. Once acquired, procedural memories are expressed rather automatically in performance and are not amenable to description (i.e., it is not easy to say “how” one reads). When talking about procedural memory or knowledge, one is likely to say, “I know how to.” In general, older adults show normal acquisition of procedural skills in both motor and cognitive domains and retain them across the lifespan. With high levels of expertise, there is often little slowing of skilled performance with age (at least until the very oldest ages), although some individual components of the skill may decline. So, for example, although the finger movements of a skilled typist slow down with age, overall typing speed is maintained because other aspects of the skill adjust (e.g., scanning further ahead in the text to be typed) [39]. Procedural memory depends on several brain regions, including the basal ganglia and the cerebellum.

5. Implicit Memory

Implicit memory refers to a change in behavior that occurs as a result of prior experience, although one has no conscious or explicit recollection of that prior experience. For example, laboratory experiments have shown that it is easier to identify a degraded stimulus (e.g., from a brief exposure or partial information) if the stimulus was seen previously, even if one does not remember the prior occurrence. This “priming” probably occurs ubiquitously in everyday life and appears relatively intact in normal aging, although there are some inconsistencies in the literature (for review, see [40]). The most extensively studied form of implicit memory is perceptual priming, which occurs in response to a perceptual cue. Perceptual priming is modality specific and depends on sensory processing areas of the brain (e.g., in the visual domain, priming involves extra-striate regions of the visual cortex). Conceptual priming, which requires semantic processing and is observed in response to a conceptual cue, is also preserved in many older adults, and has been associated with left frontal and left temporal cortical regions.

6. Prospective Memory

Much of what we have to remember in everyday life involves prospective memory — remembering to do things in the future, such as keep appointments, return a book to the library, or pay bills on time (for review, see [41]). Older adults do quite well on these daily tasks, using a variety of external aids such as calendars and appointment books to remind themselves of these activities. Certain habitual tasks such as taking medications at the appropriate times each day, however, may create difficulties for older people. For these tasks, there often are no salient reminders or cues in the environment, and so the tasks require the kinds of self-initiated activities that seem to be particularly problematic for older adults. Prospective memory may also rely on some aspect of working memory to maintain future intentions over time and likely also involves divided attention, both functions that show age-related deficits. Prospective memory and episodic memory tend not to be correlated and probably depend on different regions of the prefrontal cortex.

7. Long-Term Memory: Summary and Implications

Aging principally affects episodic memory, namely memory for specific events or experiences that occurred in the past. Although many older adults believe that their memories for remote events are better than their memories for recent events, it is likely that older memories have become more semantic or gistlike, retaining the general core information but lacking details, particularly spatial and temporal context. These older memories have often joined the realm of things that we now “know.” More problematic for older adults is remembering context or source information: where or when something was heard or read, or even whether something actually happened or was just thought about, what has been called “reality monitoring” [42]. Encoding and retrieval of these kinds of specific or peripheral details about a prior event may be particularly demanding of attentional resources, and good cues for the retrieval of such information may often be lacking. Although semantic memory is largely preserved in old age, the fact that what is retrieved from semantic memory is general knowledge, not specific detail, may contribute to the absence of age differences. The exception to this pattern might be the retrieval of a person’s name or a specific word for a specific context, both of which show deficits in normal aging. The specificity of the information to be retrieved may therefore be a critical determinant of age differences [43]. There is some suggestion that age-related deficits in memory may be reduced for emotionally arousing events or materials [38], and so emotional or personal investment in an experience may be an important variable in episodic memory in older adults. High levels of emotion or stress, however, generally have negative effects on memory.