Alzheimers Basics

The common stereotype that memory gets worse in old age actually is true for episodic memory but is not true for semantic memory or priming. Literally hundreds of laboratory studies have documented age differences in episodic memory tasks (e.g., memory for linguistic material such as words, sentences, paragraphs, and stories, and memory for visual material such as pictures, objects, faces, and scenes). Remember that these materials are presented for study and subsequently tested for retention by means of recall (e.g., write down all the words you can remember from list) or recognition tests. More recently, these age differences have been documented outside the laboratory as well (e.g., faces and names, hiding valuable objects, etc.).

These types of memory tasks are episodic because subjects are asked to remember items experienced during a personally experienced episode. In a recent review of scientific studies (1993), I found that healthy older adults remembered an average of 32 percent less than young adults. This difference occurred even though memory was tested with picture recognition, probably the easiest episodic test (age differences are typically greater for free recall tests, in which no retrieval cues are provided). Thus, healthy elderly adults can be characterized on a group basisas having a measurable, quantitative loss of retrieval from episodic memory. (Robin West and her colleagues at the University of Florida have also found that age-related decline is the primary factor in “everyday memory performance.”) The quantitative aspect is emphasized because the manner in which healthy older adults remember is not qualitatively different from younger adults.

In contrast, the episodic memory deficit in Alzheimer’s disease patients is definitely qualitative. For starters, these patients certainly remember less information. In a number of studies, dementia patients consistently recalled considerably less information compared to healthy elderly of the same age. But even more importantly, the way in which Alzheimer’s disease patients recall information is very different from healthy older adults. A couple of examples from my own research will illustrate this point. This research was conducted when was at Duke University, in collaboration with Reed Hunt (University of North Carolina at Greensboro) and Frederick Schmitt.

In one study, we asked subjects to read a number of sentences. Some of the sentences were complete, such as “The gentleman opened the door.” Other sentences were incomplete and the subject was asked to supply the missing word, as in “The teacher taught the —.” Later, we provided each person with the subject of each sentence (e.g., gentleman and teacher) and asked them to recall the object (last word) of that sentence (e.g., door for the first sentence, class, student, etc., for the second sample sentence, depending on the word generated).

Even though older adults recalled less information, both young and older healthy adults memory benefited from generating their own words. Words they thought of themselves (for incomplete sentences) were better remembered than words they read in the complete sentences. This phenomenon is known as the generation effect. The magnitude of this effect is plotted in figure as the difference between generated minus read words. The Alzheimer patients, in contrast, did not benefit from the generation effect, they recalled words poorly regardless of whether those.

Mean size of the generation effect (number of generated minus read words recalled) in young and older healthy adults and Alzheimer patients (adapted from Mitchell, Hunt, and Schmitt, 1986). Note that normal aging does not diminish the memory advantage of self-generated information, but that Alzheimer patients memory demonstrates virtually no benefit from generation.

Another episodic memory task we investigated was reality monitoring, which involves discriminating memory for internal thoughts from external perceptions or actions. For instance, as you drive to work one morning, you wonder if you actually turned off your electric coffee maker or only thought about unplugging it. Did you actually take your medicine or only think about taking it? We showed subjects all the complete and incomplete sentences they had seen earlier, with the incomplete sentences filled in with their own words. They were asked to decide which sentences were ones they had read, and which ones they had generated (completed). It is evident that healthy older adults were just slightly worse than young adults-a quantitative difference-but Alzheimer

Mean reality monitoring scores in young and older healthy adults and Alzheimer patients (adapted from Mitchell. Hunt, and Schmitt, 1986). Note that the ability to discriminate previously generated words from previously read words is only slightly impaired with normal aging, whereas Alzheimer patients have lost this ability. Patients performed much worse than either of the other groups. In fact, the Alzheimer patients’ performance was no different from chance (50 percent), which is qualitatively different from the healthy older adults. This result parallels anecdotes about Alzheimer victims who have written the same check a few times, or who have left the range on.

When patients have memory complaints or when professionals discuss memory loss, it is often assumed that memory is a singular trait. In fact, however, there is increasing evidence for several types of memory. The distinction between different memory systems is important because some systems may be affected by aging or by disease, while other systems may remain intact. At least three types of long-term memory representation (conceived by Endel Tulving at the University of Toronto) are important for understanding the differential effects of normal aging versus diseases : episodic memory, semantic memory, and implicit memory.

Episodic memory involves conscious recollection of specific events in your life that occurred in a particular time and place. What were you doing when you heard that John Kennedy was assassinated? When the space shuttle Challenger exploded after liftoff? Where did you spend Thanksgiving last year? When did you last see your spouse? What did you eat for breakfast today? All of these remembrances require episodic memory, which allows us to remember what, when, and where. Episodic memory contains information ranging from a few minutes ago to many years ago.

Semantic memory contains our vocabulary and general knowledge of the world, information that is available independent of time and context. Who was John Kennedy? What do you usually eat for Thanksgiving? What kinds of clothes should you wear to your nephew’s wedding? What is your spouse’s name? Note the difference between these questions and the episodic memory questions in the previous paragraph. In the memory laboratory, we might ask someone to name some fruits (semantic memory) or to recall the names of some fruits from a list presented earlier (episodic memory). Healthy older adults-compared to young adults do experience greater difficulty with episodic memory, but not with semantic memory. Alternatively, patients suffering from Alzheimer’s disease suffer loss of both types of memory.

Implicit memory is the most basic type of memory, as it simply requires a response in the presence of a previously experienced stimulus. At the piano or computer keyboard, our fingers seem to “know” where the keys are. When a traffic light turns red, our right foot goes automatically to the brake pedal. When we see a familiar printed word, its pronunciation is immediately available. The second time we visit a foreign country, the vocabulary and expressions come to mind with greater facility. Priming, then, is quite different from the other types of memory in that it requires no conscious recollection but does reveal the effects of prior experience (i.e., memory). In contrast to episodic and semantic memory which involve “knowing when” or “knowing what,” implicit memory has been characterized as “knowing how.”

For an implicit memory task in the laboratory, an individual, an individual might be asked to engage in word puzzles, with no mention of a memory test. Try to complete the following fragments to form words – D_ N_S_U_, AV _C_D_, T_ QU_L_, A_ R_ V _R_, G_ N_RA_I _N, and E_ E_HA_T – although it’s fairly difficult, when the whole words have been seen previously the number of fragments completed rises dramatically, providing evidence of memory. This form of memory occurs even when individuals don’t remember having seen the very same words. Even more striking. amnesics-who, by definition, have extremely poor episodic memory-perform at the same level as normals when an implicit test is used. For present purposes, this finding is of great interest because it shows that

(1) It is possible to tap information stored in memory not normally available to consciousness

(2) That separate memory systems can be differentially affected by factors such as aging and disease.

We will see evidence that implicit memory is invulnerable to the effects of normal aging and may be spared in Alzheimer’s disease as well.

Implicit memory appears to be the most basic memory system, and recent evidence suggests that it continues to function normally in old age, in amnesics, and perhaps in patients with Alzheimer’s disease. Remember that some stimulus must be presented to the subject in order to elicit a response. Either the accuracy of the response (e.g., a subject completes a word fragment) or the speed of the response (e.g., a previously seen item is named faster) reveals the functioning of implicit memory. An implicit memory task that have used involves measuring how long it takes people to name pictures of common objects, such as line drawings of a dog, trumpet, banana, or a chest of drawers, etc. Naming time is measured in milliseconds from the moment a picture comes up on a computer monitor until an individual speaks into a microphone. Only naming times for names successfully retrieved are used in the analysis. When people are asked to name pictures a second time, their naming times are faster than on the first presentation indicating implicit memory for those pictures. This is true even for pictures that people cannot consciously recollect.

Frederick Schmitt and employed this task with three groups of subjects young and old healthy adults and Alzheimer patients. We asked them to name sixty pictures, half of the pictures were presented a second time a few minutes later. Average naming times (not shown) were fastest in young adults, somewhat slower in healthy older adults, and slowest in Alzheimer patients. In spite of the group differences in overall naming speed, the three groups showed equivalent increases in name retrieval speed on the second occurrence (or repetition) of a picture. This phenomenon is called priming and is assumed to reflect the operation of implicit memory. Thus, neither normal aging nor Alzheimer’s disease seems to disrupt the functioning of implicit memory. However, many researchers are actively investigating a variety of priming tasks, because while most of these tasks reveal preserved functioning in patients with Alzheimer’s disease, some do not.

A related finding has been reported by Laura Monti and John Gabrieli and colleagues at the Rush Alzheimer’s Disease Center in Chicago. These investigators asked patients to read passages many times. Their Alzheimer patients showed normal implicit memory as evidenced by increased reading speed for identical passages on subsequent tests. This priming effect was equivalent for a group of normal elderly and the Alzheimer patients, in spite of the latter group’s very poor performance on episodic (recognition) memory tests for the same passages.

Jason Brandt and his colleagues at Johns Hopkins University have employed another priming memory task. In this task, subjects are initially shown a list of words. Later, subjects see a longer list of words, some of which are related to the original list. Subjects are asked simply to say the first word that comes to mind-thus, it is not presented as a memory task. Both Alzheimer patients and normal elderly (mostly patients’ spouses) revealed memory for the original words, in that their word associates to the new list tended to be the original words, at a greater-than-chance level. This priming memory equivalence held in spite of the Alzheimer patients’ poor episodic recall (about 37 percent of the level of healthy older adults).

Daniel Schacter, at Harvard University, published a very interesting example of intact implicit memory in a patient (MT) diagnosed with Alzheimer’s disease. Schacter took MT (an experienced “duffer” in his own words) out for a couple of rounds of golf. MT’s memory for the location of his last shot-episodic memory was quite poor (only 35 percent correct). Likewise, his episodic memory for playing golf with Schactel’ was severely impaired, as he denied having played at all when asked about it a week after the fact. In contrast, his playing ability, and knowledge of etiquette, rules, strategies, and jargon demonstrated remarkable preservation of his semantic and priming memories. In light of the research discussed above, M.T. ‘s priming memory is not surprising. His semantic memory functioning, however, is unusually good compared to the average Alzheimer patient, but is in line with his laboratory measures, which show his vocabulary and information skills to be intact. His semantic memory functioning suggests that he is only in the mild stage at this point. It would not be surprising if his golf skills (i.e., priming memory) remain intact for some time after his naming and vocabulary abilities deteriorate in the course of the disease.