In the previous section, you learned that long-term memories are stored at the preconscious or unconscious levels, and that these are called explicit and implicit memories, respectively. In this section, you will learn about the other characteristics of long-term memory. The topics of the duration and retrieval of long-term memories will be separated into two subsections: one for explicit (preconscious) memories and another for implicit (unconscious) memories.
Explicit Long-Term Memories
Duration. By definition, long-term memory codes last a relatively long time. In research situations, we typically are interested in long-term memory codes that last at least a few hours to a few weeks. When trying to estimate the duration of explicit memories, however, we run into a problem that did not occur when estimating the duration of short-term memory codes. Short-term memories are easily accessed because their memory codes are stored at the conscious level, whereas explicit memories, because their memory codes are stored at the preconscious level, can be accessed only with the proper retrieval cues. This fact can make it difficult to measure accurately the average duration of explicit memory codes: a person may not be able to retrieve an explicit memory because its memory code no longer exists (the engram has disappeared) or because the retrieval cue used to activate it was inadequate — that is, the retrieval cue was not able to activate the memory code because it was not associated closely enough with the memory. Furthermore, it’s also possible that what once was an explicit memory has become an implicit memory with the passage of time (see below).
Retrieval. A retrieval task is defined by the type of retrieval cue used to activate a long-term memory code. In research on explicit memories, two kinds of retrieval task often are used: recall and recognition tasks. In a recall task, the retrieval cue is simply a request to retrieve items of information learned at an earlier time. In other words, no specific retrieval cue is given. For example, in the section on working memory, we saw that memory researchers often use lists of numbers, words, or letters that are read aloud to study participants and then recalled immediately. Let’s say that the following word list is used:
pen cow bar man day few hot
Because this list has only seven items — which is the average capacity of the short-term store — there would be little decay of their short-term memory codes for at least several seconds, and most participants, therefore, should easily recall all seven words.
A recall task is much more difficult, however, when it involves the retrieval of memories from the long-term store. This is because no retrieval cue is provided in a recall task and, therefore, study participants must mentally construct their own cues. For example, let’s say you are asked to recall the names of all U. S. presidents from 1945 to the present. Assuming that you have, at some time or another, learned this information and stored it as a long-term memory, you must mentally search for it in the long-term store and activate the relevant memory codes. In recalling these names, many of you probably began your search by thinking of the current president (Barack Obama) and working your way backwards: George W. Bush, Bill Clinton, George H. W. Bush, Ronald Reagan …. at which point you may have begun to have trouble if you are younger than about 35 years. Perhaps you recalled that Richard Nixon and John Kennedy came sometime before Ronald Reagan, but you may not have been able to recall the remaining presidents (see the POTUS web site if you’re still having trouble).
Constructing a retrieval cue yourself (that is, searching for the information and activating the memory codes) involves complex mental processing that requires a great deal of mental effort, at least for memories that are only vaguely recalled at first. This is why many of you find essay tests to be so difficult: they are pure recall tasks that ask you to retrieve information that you learned while studying in the days or weeks prior to the test.
In a recognition task, the retrieval cue consists of items of information learned at an earlier time. In other words, individuals are asked to recognize things that he or she was exposed to in the past. For example, police often use photographic and physical line-ups, which typically are made up of five to seven people, one of which is a suspect in a criminal investigation. An eyewitness to the crime is asked to look at the faces in the hope that he or she will recognize the suspect as the perpetrator of the crime. Or to use the word-list example presented above, let’s say that a week after memorizing that list, you are given the following list of 15 words and asked to recognize the items you learned earlier:
oak pen hat cow bar man arm big gun day old few fog hot pit
People typically find it much easier to recognize items that they learned a week before than to recall those items. This is because, with a recognition task, people can use the information they are given to mentally search through their long-term store for the same information, which is much easier because the previously learned items directly activate their memory codes. Multiple-choice tests become recognition tasks when the correct choices are worded identically to information learned while studying.
Implicit Long-Term Memories
Duration. As you learned above, it can be difficult to determine the duration of an explicit memory because such memories are preconscious and, hence, require a retrieval cue to activate them. This difficulty is even more extreme with implicit memories because they are at the unconscious level and, therefore, cannot be brought into conscious awareness. The existence of an implicit memory can be determined only by its effects on behavior. Thus, it is virtually impossible to determine the duration of an implicit memory.
Retrieval. In research on implicit memories, recall and recognition tasks cannot be used because, by definition, the person cannot become directly aware of information stored at the unconscious level. Two other kinds of retrieval task often are used to activate implicit-memory codes: relearning tasks and priming tasks. In a relearning task, the retrieval cue consists of exposure to forgotten items of previously learned information that then are relearned. If the items are learned more quickly and with less effort the second time, then implicit memory codes must have been activated during the relearning. For example, people who have taken a year or two of a second language early in high school often cannot consciously recall or recognize much of it when exposed to the language again several years later in college. Nevertheless, many people can relearn more quickly what they had learned previously. The fact that they relearned the information more quickly the second time around can be explained only by the activation of unconscious memory codes stored in the long-term store.
In a priming task, a person is exposed to a set of items (usually words or images) that are forgotten over time. Later this person is given a prime for each item: a retrieval cue that consists of only a portion of the item. An implicit memory is indicated when the person responds to the prime by reproducing the original item. For example, let’s say that research participants are asked to use the following words to construct a story:
lint pine year turn bore zinc pair mine
A few weeks later, it’s likely that they will neither recall nor recognize these words because they were not asked to memorize them. Parts of these words now can serve as primes to see if the study participants formed implicit memories of the words. Let’s say that they are given the following list of primes and asked to fill in the spaces with letters to create words:
l_n_ p__e ye__ t_r_ b__e _i_c pa__ m_n_
People given such a task tend to use letters that recreate the words they were exposed to several weeks before. Again, the only way this could occur is if they formed implicit memory codes of the words they had used to create the story.
Another example of the priming effect makes use of “fragmented pictures” of objects, such as those in Figure 1. People are shown the most-fragmented pictures first and asked to identify the objects pictured. If they can’t, they then are shown the more-complete pictures in sequence (from most- to least-fragmented) until they are able to identify the objects. At a later time, they are again shown the sequence of pictures, starting with the most-fragmented, and asked to identify the objects. In this case, even patients with severe amnesia due to hippocampal damage often are able to identify the objects at an earlier stage of the sequence. This shows that, even though they no longer have explicit memories of their first experience with the fragmented pictures, they have implicit memories that, when primed with the fragmented pictures, are activated, thereby allowing them to identify the objects more easily the second time they see them (for a review, see Kihlstrohm, 1987).
In general, the priming effect refers to a tendency to respond to a stimulus (the prime) in a way consistent with an implicit memory because it is associated with the implicit memory. In our everyday lives, our mental events and behavior probably are affected by priming often and in many ways. For example, it may be that at least some of our “great ideas” — ideas that seem to be due to some mysterious power of intuition — actually are caused by priming. In short, we may have heard an idea before but no longer have an explicit memory for it. The implicit memory for the idea, however, may be activated by current events in such a way that the idea now seems to “pop into our heads.” Let’s look at an example of this taken from the history of psychiatry. A key idea in Freud’s psychoanalytic approach seems to have arisen through priming:
Freud had maintained for years an intense and tumultuous friendship with the Berlin physician Wilhelm Fliess. He frequently confided his latest ideas and insights to Fliess, and was emotionally dependent on his approval of them. When Freud announced to Fliess a momentous new insight — that every person is fundamentally bisexual — he fully expected Fliess to be amazed by the idea. Instead, Fliess responded by reminding Freud that he himself had made exactly the same discovery two years earlier and [had] told Freud all about it, and that Freud had rejected the idea. (Schacter, 1996, p. 168)
Freud eventually remembered the earlier event and realized his mistake. In many cases, however, people often do not remember the initial learning and may disagree vigorously with those accusing them of stealing a primed idea. In fact, you may have experienced such a disagreement yourself, and become very frustrated when the other person did not remember events the way you did. If you learn nothing else about memory in this chapter, you should at least learn that our memory can be very faulty.
Capacity of the Long-Term Store
It’s believed that we can store a virtually unlimited amount of information in the long-term store (Schoeke, et al., 2008). That is, over the course of a lifetime, it seems quite probable that we can never fill up our long-term stores. Everyday experience should tell you that this claim is probably true. For example, elderly people who have accumulated a great deal of information in their long-term stores over a lifetime can still learn new facts (such as the name of a person they have just met) without losing old memories. And no matter how much you have studied for the large number of tests you have had over the years, it is not likely that you will ever reach a point where you cannot learn even more information. That is, you will not fill up your long-term store with the material you have learned in your previous courses. We also can look at this issue in terms of adaptation — we would not be well-adapted creatures if we could ever get to a point where we reached the full capacity of our long-term stores. Such a human would not be able to learn new information and, hence, would not be expected to survive much longer when conditions changed.
The Encoding of Long-Term Memories
In order to encode information for transfer to the long-term store, the information must be rehearsed properly in working memory. The processing of information in working memory can occur in either a shallow (superficial) or a deep manner (Craik & Lockhart, 1972). Maintenance rehearsal involves a shallow processing of information because you simply are repeating the information over and over again while expending very little effort to transform and organize the material. On the other hand, elaborative rehearsal involves a deep processing of information because you are using much mental effort to link new information to older information already stored in long-term memory. Much research shows that the more deeply we rehearse new information in working memory, the more likely it is that it will be encoded for transfer to long-term memory.
When researchers ask study participants to memorize word lists, and then see which words are still in their long-term stores some time later, they find evidence that memories for many of these words must have been created through elaborative rehearsal. For example, if we have people memorize any one of the word lists presented in Section 5-8, and then test their memories for it a day or two later, we probably will find that, when they “remember” words that were not in the original list, these misremembered words will have a meaning that is similar to words in the original list. Let’s say that I asked you to memorize the following word list:
lint pine year turn bore zinc pair mine
If I then test your long-term memory for the list two days from now, you are likely to misremember words such as the following:
dust tree hour spin dull iron join bomb
Although you probably will remember some of the words correctly, when you do make a mistake, it is likely to be a word that has a meaning associated with the original word, which implies that you semantically encoded the information for long-term storage. In order to semantically encode verbal information, you would have used elaborative rehearsal.
Much research on long-term memory shows that elaborative rehearsal, when compared to maintenance rehearsal, leads to:
- greater transfer of information from short-term to long-term memory;
- more enduring (stable-over-time) long-term memories.
[NOTE TO MYSELF (STUDENTS-IGNORE THIS): LEVELS OF PROCESSING DISCUSSION]
Study Questions for Section 5-12
- What is the average duration of long-term memory codes?
- Why is it difficult to come up with an accurate estimate of long-term memories?
- How would you define a “retrieval task” in your own words?
- Which retrieval tasks are most often used in research on explicit memories?
- How would you define each of these types of retrieval tasks in your own words?
- What are examples of each of these retrieval tasks in your everyday life?
- Why are recognition tasks generally easier for people than recall tasks?
- Which retrieval tasks are most often used in research on implicit memories?
- How would you define each of these types of retrieval tasks in your own words?
- What are examples of these retrieval tasks in your everyday life?
- How might priming explain the experience of suddenly coming up with a great idea?
- What is the capacity of the long-term store? Why did you give this answer?
- Explicit memories of word lists typically are encoded in which way for storage in long-term memory?
Baddeley, A., & Hitch, G. J. (2010). Working memory. Scholarpedia, 5(2), 3015. doi:10.4249/scholarpedia.3015
Retrieved November 6, 2011, from http://www.scholarpedia.org/article/Working_memory
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671-684. doi:10.1016/S0022-5371(72)80001-X
Retrieved November 6, 2011, from http://picard.montclair.edu/psychology/adams/craik-and-lockhart-1972.htm
Kihlstrom, J. F. (1987). The cognitive unconscious. Science, 237, 1445-1452. doi: 10.1126/science.3629249
Retrieved November 6, 2011, from http://violet.berkeley.edu/~kihlstrm/PDFfiles/ScienceCogUncog.pdf
Lockhart, R. S., & Craik, F. I. (1990). Levels of processing: A retrospective commentary on a framework for memory research. Canadian Journal of Psychology, 44, 87-112. doi: 10.1037/h0084237
Retrieved November 6, 2011, from http://www.nclack.k12.or.us/1826201112222516913/lib/1826201112222516913/Levels_of_processing.pdf
Schacter, D. L. (1996). Searching for memory: The brain, the mind, and the past. New York: BasicBooks.
Schoeke, A., Bittlin, T., et al. (2008). Cognitive psychology and cognitive neuroscience. Wikibooks. Retrieved November 6, 2011, from http://upload.wikimedia.org/wikipedia/commons/d/d8/Cognitive_Psychology_and_Cognitive_Neuroscience.pdf
Snodgrass, J. G., Bradford, S., Feenan, K., & Corwin, J. (1987). Fragmenting pictures on the Apple Macintosh computer for experimental and clinical applications. Behavior Research Methods, Instruments, & Computers, 19, 270-274. doi: 10.3758/BF03203798
Retrieved November 6, 2011, from http://www.springerlink.com/content/77k3017662x66113/fulltext.pdf