REVIEW ARtICLE ▼ Memory, brain and aging: the good, the bad and the promising by Beth A. Ober A large body of evidence converges on the conclusion that episodic mem- ory (the recollection of personally experienced events) is the only long- Caspel van iStockphoto/Wouter term memory system that shows significant age-related deficits. More- over, the brain regions most likely to show age-related volume loss are those most critically involved in epi- sodic memory. Older adult brains may have much greater plasticity (capacity to change) than once believed; for example, neurogenesis (the birth of new neurons), increases in cognitive (including memory) performance, and High levels of mental activity, such as playing games like bridge, are predictive of improved cognitive and brain outcomes in older adulthood. increases in regional brain volume have all been shown to occur in older Skills that we have acquired over our Myth: “Alzheimer’s disease is inevita- adulthood, as a result of physical or lifetimes, most of which can be exe- ble in old age.” In fact, Alzheimer’s dis- cuted in a fairly automatic manner (e.g., ease, which accounts for about 50% of mental activity/training. The next typing, driving, reading, swimming, all dementia cases, occurs in 0.9%, 4.2% wave of research will enhance our knitting, playing a musical instrument) and 14.7%, of those 65, 75 and 85 years understanding of brain plasticity in also depend on memory. of age, respectively (Brookmeyer et al. adulthood and enable specific guide- Adults over 65 years old comprise 2007). Although the risk of Alzheimer’s 11.2% (4.1 million) of Californians. disease and other dementias increases lines for lifestyle or pharmacological Due in part to aging baby boomers, with age, a minority of older adults are treatments that optimize brain and California’s elderly population is ex- affected. memory functioning well into late pected to grow twice as fast as the total Myth: “When memory problems oc- adulthood. population to more than 8 million by cur in normal aging, they involve all 2020 (US Census Bureau 2010). Changes aspects of memory.” Only one type of in memory function, or even the per- long-term memory, known as episodic hen middle-aged or elderly ception and interpretation of such or event memory, undergoes significant Wadults discuss their health and changes by oneself or by others, can decline in normal aging; other types well-being, one of the most commonly affect day-to-day functioning and well- of long-term memory either remain expressed worries is memory problems, being in a variety of contexts, including unaffected or show improvement. including the fear of Alzheimer’s dis- family, community and workplace. Each (“Normal aging” indicates the absence ease. This is not surprising, given that and every one of our actions, interac- of dementia or other brain disease, and memory functions are at the core of our tions, musings, decisions and plans the ability to live independently.) This day-to-day intellectual and social activi- critically depend on memory. misunderstanding reflects the lack of ties. Memory enables us to know who awareness of scientific evidence for Memory myths vs. reality we are, as well as what our goals, values several separable types of long-term and beliefs are. Memory allows us to A number of myths persist about memory function, each with separable think about recent and remote events memory and aging. brain underpinnings. in our lives, and to plan for the future. Myth: “Memory abilities will decline Major types of long-term memory Moreover, memory enables us to accu- in older adulthood to a degree that will mulate and make use of vast reservoirs significantly affect day-to-day function- Cognitive scientists define short-term of information about people, objects, ing.” Rather, about 85% of adults age 65 memory as that which is available in places, social customs, language and and older do not show significant de- consciousness (e.g., the sentence that individual domains of expertise (e.g., clines in memory ability and are able to was just spoken, the phone number just birds, wine, astronomy, neuroscience). live independently. dialed), whereas long-term memory 174 CALIFORNIA AGRICULTURE • VOLUME 64, NUMBER 4 encompasses any memories that are still available after a brief period (30 to 60 seconds) of distraction or disen- gagement. Information in short-term memory will only be encoded into long-term memory and be available for retrieval after a delay if one is mo- tivated or interested in remembering it. Otherwise, it will quickly be forgot- ten. When information in short-term memory is manipulated in some way, for example in order to make linkages with information in long-term storage, or in the service of computations, this type of memory is referred to as work- Fig. 1. Increasingly complete versions of a given object are shown until the object is identified. ing memory. Source: Snodgrass et al. 1987; reproduced by permission from Life Science Associates. There is general agreement among cognitive psychology and neuroscience the party exemplify the sensory-motor scholars on the broad outlines of the type of procedural memory. 1.5 behavioral and brain aspects for the Rey Episodic recall and recognition 1 Benton following three major types of long- Free recall term memory (Schacter and Tulving Laboratory tests of episodic memory 0.5 Cued recall 1994; Schacter et al. 2000; Squire and generally involve a study phase for a set Knowlton 2000). of to-be-remembered words (or pictures 0 Episodic. Episodic memory involves or sentences), followed by a test phase Z scores the conscious recollection of episodes in which participants are asked to either −0.5 or events, along with a personal context. recall or recognize the items that they Recalling a dinner party attended last recently studied. Recall tests can be −1 week, including where the event took either “free” (e.g., “Tell me everything −1.5 place, who was there, what you ate and you remember from the list”) or cued 20–29 30–39 40–49 50–59 60–69 70–79 80–89 what the conversations were about, ex- (e.g., “Tell me all animals that you can Age groups (years) emplifies episodic memory. remember from the list”). Age-related Semantic. Semantic memory involves declines in episodic recall are well Fig. 2. Adult life span (cross-sectional) “world knowledge” in the broadest documented (fig. 2). Recognition tests performance on four episodic memory tasks. sense: languages, objects, places, spatial typically require a “yes” versus “no” “Benton” and “Rey” data are from clinical, pictorial, word-list learning tests named after relationships, social norms, facts/trivia, response as to whether an item was in the authors; “cued recall” and “free recall” data and all sorts of concepts from domains the previously studied list. Older adults are from laboratory, word-list learning tasks in which one may or may not be an show little or no deficit in memory per- created by the source author. Plotted Z scores expert. The knowledge brought to the formance for recognition, in contrast are referenced to the entire age range, with an overall mean of zero; younger adults are dinner party about the names, back- to significant deficits for recall (fig. 3). above and older adults are below zero. Source: grounds and interests of the guests is This reduction has been attributed to Park et al. 2002; © 2002 American Psychological part of semantic memory. environmental support, which reduces Association. Procedural. Unlike both episodic the effort needed for retrieval processes. and semantic memory, procedural The dissociation between recall and 25 memory is not easily accessible to con- recognition is why we often fail to recall 20 scious awareness or verbal description the name of a movie we saw or a person Recognition and evaluation. Procedural memory we met several days ago, whereas we 15 Recall includes: (1) sensory-motor skills such can easily recognize the name as correct as typing, bicycle riding and piano when it is presented to us either alone 10 playing; (2) learning and memory for or as one of several options. The older 5 procedures, sequences and rules to we are, the more often we have failures Mean number correct solve tasks and puzzles; and (3) repeti- of recall. However, when recognition 0 tion priming, an increase in speed and/ is successful for a name that cannot be 20–29 30–39 40–49 50–59 60–75 or accuracy when identifying a word recalled, this shows that the name is at Age group (years) or picture on repeated presentation as least partially available in memory but compared to its initial presentation, not readily accessible. Fig. 3. Adult life span (cross-sectional) even when the item is degraded and the The differential susceptibility performance for free recall compared to recognition on a word-list learning, episodic presentations are separated by many of recognition versus recall perfor- memory task. Data source: Schonfield and days (fig. 1). Driving skills used to get to mance in normal aging is consistent Robertson 1966. http://californiaagriculture.ucanr.org • OCTOBER–DECEMBER 2010 175 As we progress through middle and older age we acquire larger vocabularies and more extensive knowledge about the world around us. Shutterstock/camilla$$ with dual-process models of episodic that recollection is more negatively af- memory — recollection and familiarity fected than familiarity, just as recall is (Light et al. 2000). Recollection is a rela- more negatively affected than recogni- tively slow and deliberate process that tion, in aging. These two components of involves memory access for the desired episodic memory may reflect neuroana- item/episode in tandem with informa- tomically distinct memory processes, tion concerning the context in which with recollection being more dependent it was experienced, such as when or on the brain’s hippocampus, and fa- where the episode occurred, the emo- miliarity being more dependent on the tions experienced at the time, and so on.