Memory

All of us have had the experience of trying to recall the name of a particular disease and feeling as though it were on the tip of our tongue, yet being unable to do so. Then, later in the day, while working on something else, the correct term suddenly springs effortlessly to mind. Such experiences remind us not only of our great reliance on memory, but what a mysterious thing it can seem to be. In such moments, it is important that we take the opportunity to study our memories in action, because it can afford deep insight into the underlying processes of human cognition. We need not necessarily treat the mind as a black box. Even if we cannot directly observe every detail of the mental and/or neuronal processes on which memory depends, we can certainly observe it in action, and from those observations develop our understanding of how to learn more effectively.

A field of psychology called cognitive information processing has employed such empirical techniques to provide a number of important insights into the nature and operation of memory. Developed during and after World War II, cognitive information processing was based in large part on the burgeoning field of computer technology. Computers clearly have limitations as a model of human cognition, however, they also shed light on the "information processing" that goes on in the human mind. How is information input, stored, manipulated, and retrieved? A major insight concerned the fact that information passes through multiple stages or registers as it is processed. The mind does not simply absorb and store everything with which it is presented. Some of it passes by unnoticed, and what is retained is highly processed, brought into dynamic rela tion with other experiences already in memory, and assigned a meaning of one type or another. As courtroom testimony often reveals, two people can see the same event and understand and recall it in dramatically different terms.

Memory, then, is not a great monolith, but a series of levels or stages of information processing. The most primary of these is sensory memory. In sensory memory, information is available to us for a very short period of time after the stimuli have passed, perhaps but a split second. If we are to retain the information for a longer period of time, it must enter working memory. Short-term working memory is what we call consciousness. Once an event has passed, only certain features are accessible to short-term working memory. To an expert, those features are the most essential ones, such as the visual clues key to the diagnosis. I may have no clue as to what the patient was wearing, but the patient's agitation, unbuttoned collar, tremulous voice, and exophthalmos may all point clearly to a diagnosis of Grave's disease. The rest of the information, such as the color of the patient's trousers, may be lost forever.

If information is to be available to be recalled and used later, it must enter long-term memory. Interestingly, long-term memories are often the last to go in patients suffering from dementias. Memories seems to fade more quickly the more recently they were embedded, such that a patient may first forget the names of grandchildren, then children, then spouse, and then finally parents. A crucial implication of the distinction between short-term memory and long-term memory is the fact that short-term memory appears to have a finite storage capacity. That is, we can only retain a certain amount of information in our consciousness, and once we reach a certain point, we can only add more by allowing it to displace what is already there. In long-term memory, by contrast, we seem to have unlimited capacity. No one has ever managed to fill long-term memory to the point that it cannot hold another bit of information. Let us consider each of these types of memory in more detail.

A key consideration in understanding sensory memory is attention. If we are not paying attention to something, we are not likely to learn it, because it never appears in our consciousness. To take an extreme case, I am unlikely to learn much from the physiology lecture now taking place when I am across the hall in an anatomy lecture. Likewise, the physiology lecture will not do me much good if my mind is elsewhere, savoring my plans for the weekend. What do we attend to, and how can educators and learners exploit the understanding of attention to promote more effective learning? There is an old-fashioned way of getting attention that does offer some benefits; namely, simply telling people to pay attention, perhaps accompanied by a blow on the head from an eraser. However, such crude approaches are limited, and there are steps we can take to make learners want to pay attention of their own accord, rather than flogging them into it against their will.

One way to get learners to pay more attention is to show them the value of what they are learning to their own future performance as physicians. If they see how the information is relevant and how they will use it to take better care of their patients, and thereby to excel as physicians, they are much more likely to attend to it carefully. If we want medical students to pay close attention to a lecture on airway management, we can usefully stoke the flames of their interest by presenting a couple of cases where a physician's failure to understand basic principles led to a disaster for the patient. A practical educational strategy is to create opportunities to highlight such relevance. When we realize not only that we do not know something but that we really should want to know it and in fact will need to know it very soon to do our jobs, our level of interest in learning it is immediately elevated.

What other factors influence our level of attention? Key factors include the degree of similarity between competing sources of information, the difficulty of the learning task at hand, and our ability to direct our own attention. If sources of information are very similar to one another, each one is likely to make less of an impression. If, on the other hand, every other instructor for the day simply recited the lecture notes but one proceeded by asking questions of the students, the questioning approach is likely to evoke more attention from learners. When learning tasks are especially difficult, it becomes especially important to exclude competition for learners' attention. The more difficult the material, the harder students are likely to need to concentrate. In fact, concentration is one of the key learning abilities, and learning disabilities and even mental illnesses are often associated with concentration deficits. If we are going to solve a problem, we need to be able to keep our attention focused on it for a sufficient period of time.

A critical element in the development of expertise is enabling learners to move from needing to struggle to see a key feature to being able to perceive it almost automatically, with very little effort. One domain in which this ability is crucial is distinguishing between normal and abnormal findings. This applies to the assessment of mental status, the interpretation of chest radiographs, and the auscultation of the heart. It is probably an oversimplification to say that this is a simple process of template matching, comparing the finding at hand to mental models until a match is found. After all, no two physical examination findings are ever exactly alike. It is more likely that prototypes in our long-term memory are brought to mind until a closest fit is identified. As we have seen, Gestalt psychology has played an important role in enhancing our understanding of pattern recognition. For example, we organize information in relation to the context or background in which it is presented. One way to improve learning is to make sure that new information can be meaningfully situated in the context of what learners already understand.

Past experience is not always an enabling factor, and in some cases may actually impede learning. The processes of discovery and innovation, for example, require us to look at things we think we understand in new ways, to stop taking them for granted and see them anew, as if we did not understand them. One way to foster creativity is to present learners with problems that do not fit the usual categories. For example, medical students might be asked to explain how their favorite book could help them to better understand a particular patient for whom they are caring, or medical school faculty members might be asked how they would explain to local grammar-school students the nature of their research. When we think of education in these terms, we are not only transmitting information but encouraging learners to become involved in the advancement of knowledge.

If the information in sensory memory is to find its way into long-term memory, it must first undergo processing in working memory. For over half a decade, it has been recognized that most learners can hold approximately seven items in working memory simultaneously, the typical length of a phone number. There are ways to expand this capacity, however. One is called chunking. It appears that information that does not undergo further processing usually remains in working memory less than 30 seconds. After being introduced to someone, if we do not make a conscious effort to remember the person's name or it does not make an impression on us for some other reason, we are likely to forget it.

There are two techniques by which forgetting can be prevented. One is called rehearsal. Rehearsal simply means repeating the information over and over in one's mind. The process can be made even more effective by speaking the information aloud, or by writing it down. Many of us can recall a new phone number long enough to dial it, but thereafter it is lost to memory. As a means of retaining information in long-term memory, rehearsal is not terribly effective. A relatively high degree of time and effort are required, and even then, the information may not be accessible when needed, particularly when it needs to be accessed as part of a complex task. Despite these shortcomings, however, rehearsal is widely employed by medical learners.

Far better as a way of retaining information is encoding. Encoding means that we relate new information to information already in long-term memory in such way that the new information becomes more meaningful, and thus more memorable. Mnemonic devices represent a crude form of encoding, in which otherwise seemingly random anatomical facts are brought together in the form of a poem or song. An even more powerful means of encoding relies on situating new material in the context of anatomy, physiology, and pathology students already know. New information about congenital heart disease may be much easier to retain and recall if students understand it in terms of the pathway blood takes through the heart and great vessels and the various places where the flow of blood can be either obstructed or redirected. Developing such a foundation takes more time than a crude mnemonic device, but eventually pays much bigger dividends in terms of understanding.

Another technique for helping learners process information effectively in working memory is categorization. When students examining a patient encounter a finding such as a pelvic mass, it helps if they can formulate in their own minds a list of the different organ systems from which it might arise. Is it gastrointestinal, urinary, reproductive, musculoskeletal, and so on? Then they can make further use of the categories to assess the likely point of origin. For example, if the patient reports seeing blood on their underwear, is it coming from the urethra, the vagina, or the rectum?

Encoding is not a passive process, and learners cannot do it in their sleep. Thus expecting students to do a lot of effective encoding while they sit and attempt to pay attention to a boring lecture is not likely to be effective. Instead, we need to adopt educational strategies that encourage students to be actively engaged in encoding the new material. Lectures are not necessarily ineffective, as long as the lecturer keeps learners actively engaged by asking questions. When learners are reading, they can achieve some of the same objectives by asking questions of themselves. Our model should not be that of a person standing in the shallows of a beach, letting the waves roll over him. Rather, we should see learners, and encourage learners to see themselves, as active explorers, posing questions and solving problems. New educational technologies can be helpful in this regard, by building question and answer and exploration into the learning model.

There is no single model to explain long-term memory, but one key distinction in different models is that between word-based representation and image-

based representation. When we store an image in memory, we do so not as an exact copy of the image we saw, but as an inexact representation in which some features are accentuated and others are suppressed or even entirely omitted. We may be able to describe with a high degree of accuracy the appearance of the heart on a patient's chest radiograph, because the diagnosis was congestive heart failure, but have difficulty saying what the stomach bubble looked like, because it did not seem relevant. Happily, we can enhance our recall of images in part using verbal representations, and we can use images to enhance our verbal recall. For example, in attempting to recall the steps in the Krebs cycle, some learners may call to mind a diagram, and others may recall the steps in words. In most cases, however, we store the information in both forms, and each can facilitate the other.

This highlights one of the most fascinating problems in learning theory, retrieval of information from long-term memory. As a general rule, information becomes easier to retrieve from long-term memory the more times it has been retrieved in the past. Why would information be retrieved? In some cases, such as a phone number, it is simply retrieved to be recited and then filed away again. In other cases, however, it is retrieved to be used in solving a new problem. Generally speaking, information will be available to a greater degree in problem solving when it has been retrieved to solve problems in the past. Hence tests that ask learners not only to recall information but to use it to solve problems of the sort they will encounter in real clinical practice generally offer a greater learning opportunity. For example, examinations could present learners with new information and ask them to interpret it or use it to solve a problem using what they already know.

There is a difference between recognition and recall. Recognition involves a lower level of recollection, simply asking learners to know at what they are looking. Recall, by contrast, asks learners to bring information to generate their own answers. Most multiple-choice examinations focus on recognition, asking learners to choose the correct answer from a list of alternatives. In recall, however, they must not only select the correct answer, but formulate it for themselves. To facilitate recall, it is beneficial to help learners employ the same cues to both encode and retrieve the information. In practice, by providing multiple different cues for encoding, educators can increase the probability that at least one will be available in a real-life situation where the information needs to be retrieved. Likewise, it is helpful if the learner's physical environment and state of mind are similar in both encoding and retrieval. If information needs to be retrieved when the learner is standing, it may make sense to learn that information in a standing position. This is part of the reason the military tends to teach combat principles in high-stress situations, because that is the context in which recruits will need to employ them.

Why do we lose or forget information? The model of cognitive information processing offers a systematic approach to this problem. One problem can be the failure to encode information effectively. If information never makes it into long-term memory in the first place, then it will not be available for use later. To avoid this, learners need to be actively engaged in learning, being asked or asking themselves questions and examining what they are learning from multiple points of view. Another problem is the failure to access encoded information. This can be prevented in part by ensuring that information is encoded in multiple forms. For example, both verbal and imaginal systems can be used to encode the same information. A third problem is interference, in which other information gets in the way of what we are trying to learn. This can be avoided by reducing distractions as much as possible, and building a curriculum in which the parts reinforce rather than interfere with one another.

If we aim to excel as educators by helping our students excel as learners, we need to understand the learning process, of which memory is a key constituent. By availing ourselves of the insights of cognitive psychology, we can gain deeper insight into key learning processes such as attention, encoding, and recall, and thereby foster more efficient and more effective learning.

0 0

Post a comment