1. What is the average duration of long-term memory codes?
2. Which retrieval tasks are most often used in research on explicit memories?
3. What are examples of these retrieval tasks in your everyday life?
4. How are recall tasks similar to recognition tasks?
5. How do recall tasks differ from recognition tasks?
6. Which retrieval tasks are most often used in research on implicit memories?
7. What are examples of these retrieval tasks in your everyday life?
8. How are relearning tasks similar to priming tasks?
9. How do relearning tasks differ from priming tasks?
10. What is the capacity of the long-term store?
11. Explicit memories of verbal material typically are encoded in which way for storage in long-term memory?

Figure 1 presents an illustrated summary of what you have learned so far about sensory memory, working memory, and long-term memory.

Figure 1. The Components of the Memory System.

The major theories of forgetting from the long-term store are listed at the bottom of Figure 1. These will be discussed later in this section after we first return to a discussion of the hippocampus and its role in the formation, storage, and retrieval of explicit memories.

What Does Organic Amnesia Tell Us About Long-Term Memory?

Some of the earliest evidence in support of the claim that the hippocampus is important in memory formation can be found in the case study of Henry M. — a case that you first learned about in Section 4-2.  In 1953, the hippocampus (as well as the amygdala and some neighboring structures in the cortex) on each side of Henry's brain were removed because of very severe epilepsy that had not responded to medication. As soon as he woke up from the operation, however, Henry exhibited severe anterograde amnesia involving explicit long-term memories. To be specific, he could remember recent events for only about 30 seconds (Hilts, 1995). For example:

• He was unable to learn his way around the hospital after the operation.
• Soon after reading something, he could remember neither what he had read nor that he had read it.
• After moving to a new house, it took him eight years to learn how to get from one room to another.
• He never learned to find his way back to the house from a distance of more than two blocks.
• When he was moved into a nursing home in 1980, he never learned where he lived or who cared for him.

Henry’s memory problems involved mainly the ability to form explicit memories, especially those involving life events. On the other hand, he seemed to have little or no trouble forming new implicit memories, especially those involving new behaviors and skills. For example, Henry learned to read words written backwards, to solve particular puzzles, and to walk to the room in which he was tested each year at MIT. The fact that he could perform these tasks correctly demonstrates that Henry had implicit memories for the corresponding skills. Nevertheless, he did not remember that he knew how to perform these tasks: he had not formed explicit memories of having learned the skills. For example, when walking to the testing room, he would state that he did not know where he was going or why he was walking in that direction.

Although Henry's anterograde amnesia seemed complete, research in later years showed that he was able to recall explicit memories of some events that had occurred after 1953:

For example, Henry seems to remember something about a President Kennedy and an assassination, even though the killing occurred ten years after Henry’s surgery. But ... you couldn’t say he remembers the assassination of John Kennedy exactly. ... If asked about an assassinated president he will remember McKinley, and recall an attempt on Franklin Roosevelt’s life. If Dallas is mentioned, he may say, “Kennedy!” Asked for a first name, he may say Ted. Then, asked to reconstruct the Kennedy assassination, he will speak of an assassin who was trying to shoot the president but someone else got in the way; this is a description of the killing of Chicago mayor Anton Cermak, who was killed in 1933 by a bullet intended for Franklin Roosevelt. ... It seems as if there is some leakage from the world into his memory, some vague impressions which have become something like memories. (Hilts, 1995, p. 113)

To a small degree, Henry was able to develop semantic memories, which are explicit memories that consist of general knowledge about an object, event, activity, or situation. For example, fill in the following blanks: (a) Abraham Lincoln was the ___th president of the United States; (b) ______ was the first person to step on the moon; (c) William McKinley was shot in the city of ______. When you study for a test, you are trying to develop semantic memories. You probably also have semantic memories of your social security number, your telephone number, your birth date, and your street address.

When you remember events from your own life, on the other hand, you are retrieving episodic memories, which are explicit memories of events, activities, and situations that include the memory of one's self participating in them. In other words, episodic memories are memories of personal experiences. For example, memories of what you did ten minutes ago or of your first day of kindergarten are episodic memories. When you tell someone what you had for dinner last night, who was there, and what time it was at, you are recalling an episodic memory as long as you remember your own participation in the dinner.

People with damage to their hippocampi have more trouble storing and retrieving episodic memories than they do with storing and retrieving semantic memories. A good example of this can be found in Oliver Sack's (1995) case study of Greg, first described in Section 1-1. In 1991, Sacks took Greg to a concert featuring Greg's favorite band, the Grateful Dead. Because Greg’s amnesia extended back to the late 1960s, he had no semantic or episodic memories of music performed by the Grateful Dead since that time. At the concert, Greg seemed to enjoy himself but was confused when the "new" music was performed:

The first half of the concert had many earlier pieces, songs from the sixties, and Greg knew them, loved them, joined in. His energy and joy were amazing to see, he clapped and sang nonstop, with none of the weakness and fatigue he generally showed. ... But the second half of the concert was somewhat strange for Greg: more of the songs dated from the mid- or late seventies and had lyrics that were unknown to him, though they were familiar in style. He enjoyed these, clapping and singing along wordlessly, or making up words as he went. ... The newer songs [from 1980 and after] went far beyond any development that he could have imagined, were so beyond (and in some ways so unlike) what he associated with the Dead, that it “blew his mind.” It was, he could not doubt, “their” music he was hearing, but it gave him an almost unbearable sense of hearing the future. (Sacks, 1995, pp. 75-76)

By the next morning, Greg no longer remembered having been at the concert or having heard any of the newer songs: his episodic memories for the concert had decayed from working memory. Nevertheless, he sang some of the new songs he had heard for the first time the night before! That is, he had developed semantic memories for them. This suggests that the hippocampi (along with the amygdala and parts of the cerebral cortex) are more involved with the encoding and storing of new episodic memories than of new semantic memories.

In Figure 2, a summary of the components of the long-term subsystem are illustrated.

Figure 2. The Components of Long-Term Memory

Study Questions

12. What kind of amnesia were Henry M. and Greg suffering from?
13. How is anterograde amnesia similar to retrograde amnesia?
14. How does anterograde amnesia differ from retrograde amnesia?
15. If you know people suffering from amnesia, which type(s) of amnesia are they experiencing?
16. Which structure in the limbic system is of central importance for forming and retrieving long-term memories?
17. Were Henry M.'s and Greg's anterograde amnesia complete?
18. How are semantic memories similar to episodic memories?
19. How do semantic memories differ from episodic memories?
20. What are some examples of semantic memories in your everyday life?
21. What are some examples of episodic memories in your everyday life?
22. Which type of long-term memory do people with damage to their hippocampi have the most difficulty storing and retrieving?
23. What are the major components of the long-term memory subsystem?
24. What is an example from your everyday life of each component?

How is the Hippocampus Related to Long-Term Memory?

In humans, although it is known that the hippocampi are involved in the encoding, storing, and retrieving of explicit memories, there are several competing theories that attempt to describe the specific role(s) of the hippocampi in these processes: the consolidation theory, the retrieval theory, and the spatial theory. The consolidation theory states that the hippocampi are needed for memory consolidation — a term that refers to a set of mental processes that lead, over time, to the establishment of durable explicit-memory codes in the cerebral cortex. According to this theory, the hippocampi store a memory code until a corresponding engram has developed in the cerebral cortex — a process that may take from several days to several months. Once the cortical engram has developed, the hippocampi no longer are needed for storage or retrieval. This theory explains why Henry M. did not lose explicit memories for events before the surgery: his hippocampi allowed him to develop cortical engrams for events that occurred up to a short time before their removal. After their removal, however, Henry no longer could store explicit-memory codes long enough to allow their consolidation in the cortex.

In Section 1-2, it was stated that, in order to determine whether or not a theory is a good one (that is, whether or not it is likely to be true), researchers need to do two things:

• make a prediction about what should be observed if the theory actually is true;
• make the relevant observations.

The consolidation theory can be tested by making observations relevant to the following prediction: destruction of the hippocampi will cause (a) severe and permanent anterograde amnesia[∂], and (b) severe and permanent retrograde amnesia[∂] for events experienced up to several months before the damage. Observations of people with severe damage to their hippocampi, however, have demonstrated that the consolidation theory has two major limitations. First, the hippocampi do not seem to be solely responsible for the consolidation of explicit memories: both Henry M. and Greg showed incomplete anterograde amnesia, especially with respect to semantic memories. Second, the hippocampi seem to be involved in the storage and/or retrieval of even fully consolidated memories: Jimmie G. (Section 4-1) had retrograde amnesia that extended back to about 1945, even though his hippocampi were not damaged until the 1960s.

The retrieval theory states that one of the primary functions of the hippocampi is to search for and find memory codes already consolidated in the cerebral cortex. For example, according to this theory, when you are searching for an answer to a question on a test, your hippocampi are important for “finding your way around your cortex ” until you detect the information you want. In this case, you can think of the hippocampi as being like "memory tour guides” directing you to stored information:

If we need the name of a bird we search in restricted parts of neural space for birds [sic] names, pushing incorrect bird names and even the names of other flying creatures out of the way as we go. The idea that we have a structure dedicated to finding things in our brain provides an explanation for curious slips of the tongue, such as when we are thinking of the name of one person but accidentally and inappropriately blurt out the name of another. It might also account for the curious symptoms of deep dyslexics[∂]; when given a word like “bird” to read, they may say “butterfly.” Possibly these errors indicate that the finding system has made a mistake. (Kolb & Whishaw, 1996, p. 445).

This theory does not rule out the possibility that, to some extent, people with hippocampal damage are able to encode and store new explicit memories. Instead, the theory states that they experience difficulties finding and retrieving these memories. In fact, we saw that both Henry M. and Greg were able to remember vaguely some events that had occurred after their hippocampi had been destroyed. This theory also is consistent with the finding that retrograde amnesia can extend back several decades before the occurrence of hippocampal damage: the damaged hippocampi no longer can find the memory codes. On the other hand, retrieval theory doesn't explain why memories older than this still can be easily retrieved.

The spatial theory states that the hippocampi are important for encoding and storing spatial information. Studies in rats have shown that cells in the hippocampi are activated when they sense objects in particular spatial relationships that they have experienced before, such as when they are in a familiar location or when they are travelling in a particular direction. In one fascinating study (Louie & Wilson, 2001; see this report; also see reports in Lindsay, 2001 and Lucentini, 2001), rats were trained to find their way through a maze while heightened activity in particular hippocampal cells was observed. During dream sleep on following nights, these same cells became active. The researchers stated they were able to tell where the rats were in the "dream maze." In short, studies such as this one demonstrate that the hippocampus is important for encoding and storing cognitive maps, which, as defined in Section 3-4 under the topic of "latent learning," is a mental representation of the layout of a location in the environment. Studies with various species of mammals have shown that intact hippocampi are required for the successful completion of spatial memory tasks. For example, if you remove the hippocampi of a monkey, show it the location of a "treat," and then wait a few minutes, the monkey generally is unable to remember where the treat was placed. Thus, when their hippocampi are removed, mammals generally are unable to form spatial memories (Kolb & Whishaw, 2003).

Without fully functioning hippocampi, humans also probably cannot remember explicitly where they have been or how to get where they want to go, although they may (over long periods of time) be able to form implicit memories for these tasks, as in the case of Henry M. Brain-imaging studies show that people with intact brains show greater activity in their hippocampi when navigating and remembering directions. For example, one study (Maguire, et al., 1997) measured hippocampal activity in London taxi drivers who had been asked to recall complex routes around the city. The taxi drivers showed increased activity in their right hippocampi while performing this task, which led the researchers to conclude that the right hippocampus is involved in "the processing of spatial layouts established over long time courses." In a later study (Maguire, et al., 2000), this research group found that the posterior (back part of the) hippocampus was larger in London taxi drivers than in others, and that the posterior hippocampus was largest in taxi drivers with the most experience. They concluded that the "posterior hippocampus stores a spatial representation of the environment and can" grow larger in those who must often navigate in and remember complex routes. In fact, many studies have shown that the hippocampus is able to produce new cells (for example, Gould, et al., 1999).

In summary , it seems that the hippocampus is responsible for a number of memory functions. In general, the hippocampus influences the consolidation and/or the retrieval of explicit memories, especially episodic memories and spatial memories. One interesting application of research on the hippocampus has been the development of an artificial hippocampus that, one day, may replace severely damaged hippocampi in humans (Sandhana, 2005).

Study Questions

25. How would you describe "memory consolidation" in your own words?
26. What is the consolidation theory of hippocampal functioning?
27. How was this theory tested and what do the results suggest about the theory?
28. What is the retrieval theory of hippocampal functioning?
29. What are its strengths with respect to the consolidation theory?
30. What is one problem with the retrieval theory?
31. What is the spatial theory of hippocampal functioning?
32. How is dreaming and cognitive maps related to this theory?
33. Why were London taxi drivers used to test this theory?
34. What were the results of the taxi-driver studies and did they support the spatial theory?
35. Why were the posterior hippocampi of the taxi drivers larger than those of others?

How Do We Encode, Store, & Retrieve Episodic Memories?

We have distinguished between two types of explicit memories: semantic and episodic memories. The hippocampus seems to be especially important for the encoding, storing, and retrieving of episodic memories. During the 1930s and 1940s, a neurosurgeon by the name of Wilder Penfield and his colleagues performed research that, at the time, seemed to indicate that engrams for episodic memories were located in a specific part of the cerebral cortex. Penfield's goal was to link particular mental functions (such as visual perception) with activity in particular cortical areas. In order to achieve this goal, he and his colleagues activated areas throughout the cerebral cortex in patients undergoing neurosurgery. They found that when areas within the temporal lobes of the cortex were stimulated, a small proportion of their patients (less than 4%) reported experiences that, to the patients, felt like episodic memories from long ago. These experiences were so full of vivid perceptual details that the patients felt as if they were reliving the events:

One of Penfield’s patients was a young woman. As the stimulating electrode touched a spot on her temporal lobe, she cried out: “I think I heard a mother calling her little boy somewhere. It seemed to be something that happened years ago ... in the neighborhood where I live.” Then the electrode was moved a little and she said: “I hear voices. It is late at night, around the carnival somewhere, some sort of travelling circus. I just saw lots of big wagons that they use to haul animals in.“ (Blakemore, 1977; quoted in Baddeley, 2004, p. 114)

Penfield and his colleagues believed that they were accessing actual episodic memories — memories containing all the details of the original events. They reasoned from this that episodic memories are stored in the temporal lobes of the cerebral cortex and that they are encoded in the form of detailed reproductions of the original perceptual experiences. This research provided what seemed to be conclusive evidence for what we may call the reproduction theory of explicit memories: for each episode in our lives, we encode and store an exact reproduction that is accurate down to the smallest perceptual detail. This theory, however, is not supported by any other scientific evidence. In fact, the best evidence available supports the claim that we forget almost everything that has ever happened to us because of the decay of memory traces in the sensory and short-term stores, and the displacement of information in the short-term store as new information pushes out older information. What ends up in the long-term store from any one episode in our lives is a very small fraction of the information initially processed in sensory memory.

So, if Penfield and his colleagues weren't retrieving episodic memories, then what were their patients experiencing? In order to answer this question, there are two things we need to keep in mind. First, only a small proportion of patients reported these experiences. If episodic memory codes are located in the temporal lobes, then most patients should have retrieved vivid episodic memories when the lobes were stimulated. Second, activation of sensory areas of the brain typically causes hallucinations[∂] — vivid perceptual experiences of objects or events that are not actually occurring. In the case of Penfield's patients, it is likely that activation of their temporal lobes caused them to experience hallucinations coupled with the illusory feeling that the hallucinated events had actually happened to them. In fact, it now is generally agreed that these patients were not experiencing memories of past episodes. Instead, they were experiencing very complex hallucinations. Most research suggests that no specific area of the cortex stores episodic memory codes. Rather, episodic memory codes probably are distributed throughout the cerebral cortex.

If episodic memory codes are not reproductions of past events, then why are we seemingly able to remember so many details when we retrieve these memories? In fact, we often are able visualize in detail the locations in which events took place and the people who were there. How is this possible if the memory codes don't include these details?

Reconstruction of Episodic Memories
When we encode in working memory an event that is to be transferred to episodic memory, we attend to and elaborate only parts of the event. For example, let’s say that you are watching a frightening movie about a serial killer. During the movie, you are much more likely to encode details that are relevant to the main theme of the movie — such as the large knife that the killer used to mutilate his victims — than you are to encode details that were irrelevant to this theme — such as the fact that the room in which he imprisoned his victims had a radio in it. Later on, if asked whether a radio was present in the room, you probably will not remember this detail because you did not attend to it. If asked to describe a particular scene from the movie, you will piece together the bits of information you encoded and construct something that contains many gaps. In order to form a complete memory — one that can be consciously recalled and described — you must fill in the gaps with what you think probably happened during the scene. This process of "filling in the gaps," referred to as reconstruction, is performed unconsciously (without your awareness). In general, the memory codes that form the basis of episodic memories consist of a small number of “memory fragments” — bits of encoded information that consist of only a few details.

In trying to understand the reconstruction theory of explicit memories, it may help to think of the retrieval of an explicit memory as being similar to the reconstruction of a complete dinosaur skeleton from a small and incomplete set of fossilized bone fragments:

For the paleontologist, the bone chips that are recovered on an archeological dig and the dinosaur that is ultimately reconstructed from them are not the same thing: the full blown dinosaur is constructed by combining the bone chips with other available fragments, in accordance with general knowledge of how the complete dinosaur should appear. Similarly, for the rememberer, the engram (the stored fragment of an episode) and the memory (the subjective experience of recollecting a past event) are not the same thing. The stored fragments contribute to the conscious experience of remembering, but they are only part of it. (Schacter, 1996, pp. 69-70)

When a set of fossilized dinosaur bones are found, it typically consists of only a small portion of the original skeleton. In order to reconstruct a complete skeleton from the small number of bones, paleontologists must use their general knowledge of what dinosaurs probably looked like — knowledge that they have accumulated since 1841, when the first dinosaur fossils (found in 1822) were recognized as being from species that had existed ages before. Because we now have so much knowledge about dinosaurs, our reconstructed versions probably are very accurate. Nevertheless, over the years, the builders of reconstructed skeletons often have made mistakes because their knowledge of the original creatures is limited by the fact that the last dinosaurs died about 65 million years ago. In a similar way, we use our general knowledge of what usually happens, or what we think must have happened, when we reconstruct an explicit memory from the small number of encoded and stored memory fragments.

When an explicit-memory code is activated by a retrieval cue, we use the knowledge that we have accumulated over our lifetimes to fill in the large gaps in the memory code (see Figure 3).

Figure 3. Reconstruction of an Explicit Memory

This set of reconstructive processes leads to a remembrance that is very much like a reconstructed dinosaur: it probably is accurate in broad terms, but often is wrong in details. Furthermore, reconstruction occurs unconsciously so that we are unaware that we have added information to the retrieved memory that was not included in the memory code. Loftus and Ketcham (1991) concluded that the process of memory reconstruction introduces inaccuracies into each and every long-term memory that we retrieve:

Memories don’t just fade, as the old saying would have us believe; they also grow. What fades is the initial perception, the actual experience of the events. But every time we recall an event, we must reconstruct the memory, and with each recollection the memory may be changed—colored by succeeding events, other people’s recollections or suggestions, increased understanding, or a new context…. Truth and reality, when seen through the filter of our memories, are not objective facts but subjective, interpretive realities. We interpret the past, correcting ourselves, adding bits and pieces, deleting uncomplimentary or disturbing recollections, sweeping, dusting, tidying things up. Thus our representation of the past takes on a living, shifting reality; it is not fixed and immutable [unchangeable], … but a living thing that changes shape, expands, shrinks, and expands again, an amoebalike creature with powers to make us laugh, cry, and clench our fists. Enormous powers — powers even to make us believe in something that never happened. (p. 20)

Let’s take a real-life example of the retrieval of an episodic memory that shows the types of inaccuracies introduced when reconstructing a complete memory from a small amount of encoded information. In 1968, Jack Hamilton, a pitcher for the California Angels, threw a fast ball that hit Tony Conigliaro, an outfielder for the Boston Red Sox, on the left side of his face. Years later, Hamilton recalled the event:

“I never hit a guy that hard.... He [Conigliaro] went right down, he just collapsed.... I’ve had to live with it, I think about it a lot,” Hamilton, now fifty-one years old and the owner of a Midwestern restaurant chain, told The New York Times after hearing about Conigliaro’s death [in 1990]. “Watching baseball on TV, anytime a guy gets hit, I think about it. It was like the sixth inning when it happened. I think the score was 2-1, and he was the eighth hitter in the batting order. With the pitcher up next, I had no reason to throw at him.” It was a day game, Hamilton recalled, because he remembered visiting Conigliaro in the hospital later that afternoon. After the accident he remembered wondering whether or not to return to Fenway Park later that year for another series of games; eventually he decided to make the trip. (Loftus & Ketcham, 1994, p. 76)

Although this reconstructed memory had one element of truth to it (Conigliaro was hit by a baseball pitched by Hamilton during a game played in Boston), the rest of it was almost completely inaccurate. Conigliaro was hit by Hamilton's pitch in the fourth inning, not the sixth; it happened during the final road trip to Boston, not earlier in the season; the game was at night, not during the day; the score was 0-0, not 2-1; Conigliaro was the sixth batter in the batting order, not the eighth and, therefore, the pitcher would not have been up next. Hamilton had recalled this memory hundreds, perhaps even thousands, of times over the 22 years since it had first occurred. During each reconstruction, he filled in the gaps in his memory code with more and more inaccurate information until most of the details of the story had changed. Nevertheless, the general theme of the story (that Tony Conigliaro had been hit hard by one of Hamilton’s pitches at Fenway Park) stayed the same.

Thus, the reconstruction of explicit memories (both semantic and episodic) is one reason why we forget long-term memories, especially those that were initially encoded and stored many years ago. In the rest of this section, we will look more closely at why we forget explicit memories.

Study Questions

36. How did Penfield and his colleagues study the location in the brain of explicit memory codes?
37. What did they conclude about the location of these memory codes?
38. What is the reproduction theory of explicit memories?
39. What do modern memory researchers believe regarding the location in the brain of explicit memory codes?
40. What do modern neurologists and memory researchers think happened when Penfield stimulated the temporal lobes and his patients seemed to retrieve episodic memories?
41. What is the reconstruction theory of explicit memories? (In your answer, please describe the nature of memory codes and what happens during the retrieval of explicit memories.)
42. How does the reconstruction theory explain the forgetting of explicit memories?
43. What does the reconstruction theory imply about the accuracy of our retrieved explicit memories?

Why Do We Forget Explicit Memories?

In this final part of Section 4, four major theories of why we forget explicit memories are discussed: reconstruction theory, encoding-specificity theory, interference theory, and defensive theory.

Reconstruction Theory
As just discussed, reconstruction theory states that forgetting occurs because the process of retrieval from the explicit-memory store introduces inaccuracies. Whenever we retrieve an explicit memory, we change its memory code to some extent by filling in gaps in the memory code with information obtained from other sources. The changed memory code will be resturned to the explicit-memory store until the next retrieval. Over time, the memory may become so transformed that it bears little resemblance to the information originally learned.

A study that provided supporting evidence for reconstruction theory was performed by two cognitive psychologists, Ulric Neisser and Nicole Harsch (1992). These researchers decided to take advantage of a highly publicized tragedy, the explosion of the space shuttle Challenger, to study something called “flashbulb memories.” Flashbulb memories are defined as episodic memories of events that, because of their emotional intensity, are encoded in vivid detail that changes very little over time. The argument behind the concept of flashbulb memories was that, because very distressing events are attended to closely and, therefore, encoded in great detail, the events are "burned" into our memories, thereby resulting in an almost photographic memory code. Thus, many people claimed to remember exactly what they were doing and whom they were with when they first heard about the assassination of President Kennedy or the sinking of the Titanic. In order to test such claims, Neisser and Harsch, the day after the explosion of the Challenger in 1986, asked students in an introductory psychology course to fill out a questionnaire that asked them to describe the situation in which they first heard the news of the disaster: where they were, how they found out, who was with them, what they were doing, and what time it was. Since the event had occurred only 24 hours before they filled out the questionnaire, their episodic memories should have been very accurate.

Three years later, Neisser and Harsch (1992) gave these students a new copy of the questionnaire and asked them to answer again the questions. They also were asked to rate how confident they were in the accuracy of their answers. The researchers compared the first set of answers to the second set for each student. They discovered that only 3 of the 44 students (7%) showed perfect recall. Thirty students (48%) recalled memories that contained varying amounts of accurate and inaccurate details. And 11 students (25%) recalled memories that were completely inaccurate. For example, one student provided the following answer 24 hours after the explosion: “I was in my religion class and some people walked in and started talking about [it]. I didn’t know any details except that it had exploded and the schoolteacher’s students had all been watching which I thought was so sad” (quoted in Ofshe & Watters, 1994, p. 39). Three years later, this same student provided a very different answer: “When I first heard about the explosion I was sitting in my freshman dorm room with my roommate and we were watching TV. It came on a news flash and we were both totally shocked. I was really upset and I went upstairs to talk to a friend of mine and then I called my parents” (p. 39). Neisser and Harsch found that over 90% of the students had memories that contained at least one major inaccuracy. What is even more surprising is that those with completely inaccurate memories were just as confident in the accuracy of their memories as were those with completely accurate memories! For example, the student quoted in this paragraph had complete confidence in the accuracy of her three-year-old memory — a memory that was false in all details.

One semester after the second questionnaire was given, Neisser and Harsch (1992) asked the students to look at their answers to both questionnaires. The researchers expected that those who had misremembered the original event would realize their mistake after reading their answers to the first questionnaire and, after being reminded of what actually had occurred, then would remember the event accurately. But not one student did so. Although most were upset by the differences between the two sets of answers, their false memories did not change. They still felt certain that they remembered the event accurately, even after being shown incontrovertible evidence that this was not the case.

Encoding-Specificity Theory
Encoding-specificity theory states that forgetting is due to an inability to retrieve stored information when the retrieval cue is dissimilar to the form in which information was encoded for storage. As stated earlier, we probably always need some sort of retrieval cue to get information out of long-term memory. In order to serve adequately as a retrieval cue, the stimulus must have some similarity to the way in which we have stored that information in the long-term store. For example, a memory that is encoded in a tactile form (such as a memory for how an unseen bump on your back feels) is not likely to be retrieved by a cue that is in visual form (such as a picture of that same bump). The encoding-specificity principle states that the ability of a retrieval cue to facilitate (aid) the retrieval of a stored memory depends upon the degree to which it is similar to the memory code. Let’s say that there is a particular woman in your class whom you know by sight. In this case, your long-term memory for her is encoded in a visual form (when you think of her, you see an image of her that you have retrieved from your long-term store). Now, let’s say that you get assigned to a group with this person and she calls you in order to discuss the project. In this case, you may not know who she is because her voice, which is serving as the retrieval cue, is an auditory stimulus, not a visual one. In other words, because you have encoded your memory of her in a visual form, her voice is not an adequate retrieval cue for activating this memory. In a sense, you have “forgotten” who she is.

There are more subtle examples of forgetting due to a mismatch between retrieval cues and memory codes. Let’s say that you have studied for a test primarily by memorizing the definitions of terms word-for-word. Rote memorization requires that you memorize the sounds of the words. Your memory codes for such definitions, therefore, would involve phonemic encoding. On a multiple-choice test, the retrieval cue is the correct answer (presented as one of several choices). If the correct answer is worded exactly as it was presented in the textbook, you should have no difficulty picking it because the form of the retrieval cue is identical to the form in which you encoded it. On the other hand, if the instructor gives an answer that is worded differently from the original definition — an answer that means the same thing — then the form of the retrieval cue is not similar enough to your memory code: the retrieval cue involves a semantic (meaning) code but you have a phonemic memory code. After the test, you may even complain angrily: “I don’t know where he got those questions! I know the answers weren’t anywhere in the book or in my notes!” In a sense, you are correct; but the problem was in the way you studied instead of in what you studied. If, during your studying, you had encoded information by forming more complex memory codes for the information (codes formed by using elaborative rehearsal), you would have been better able to choose the correct answers on the test. Thus, the encoding-specificity principle shows that how you encode information in working memory for transfer to long-term memory will affect your ability to retrieve this information later. In order to do your best on tests, you need to elaboratively rehearse information in working memory so that you can form complex semantic codes for your explicit memories. Not only do semantic codes allow for a more enduring long-term memory, they also allow a greater number of retrieval cues to activate any particular memory.

Interference Theory
Interference theory states that forgetting is due to an inability to distinguish several separate memories that are similar in some way. In other words, when separate items in memory share an important similarity (for example, several separate phone numbers or several separate names), we may begin to confuse the separate items and, hence, will have more difficulty extracting any particular item from the long-term store. There are two ways that interference can cause forgetting of explicit memories: (a) old information interferes with the memorizing of new information; (b) new information interferes with our memories for old information. For example, I often have trouble learning the names of my present students because I keep getting them confused with the names of former students (the older names make it hard for me to remember the new names) and I often have difficulty remembering the names of former students because I get them confused with the names of my present students (the new names make it hard for me to remember the older names). In general, interference is greatest when the competing items are similar in some way—the more similar, the greater the interference.

There are limits to interference. In fact, we often find that learning similar information in two different courses can help you to better remember the information. Such facilitation of learning is probably much more common than interference. Why isn’t interference occurring in this situation? Well, it is, but it depends on what we are testing for. We probably would see interference if I asked you in which of these two courses you had learned a particular item of information. But of course, no one asks you such questions on a test. Your teachers simply want you to learn the material, not also remember in which class you learned it. Thus, you should remember that interference studies are somewhat contrived and artificial: they involve rote memorization of very similar and simple material.  In general, it is probably safe to say that interference is an important cause of forgetting only when we learn similar items of information that we need to keep separate in our minds for later retrieval. For example, parking in the lot outside of school each and every day can cause problems when you are trying to remember where you parked today: you probably keep getting this confused with where you parked during the past week. Nevertheless, parking in the same lot each day also facilitates your learning of other material. For example, you probably have learned a great deal about where you are most likely to find an open space at a particular time.

Defensive Theory
Defensive theory states that forgetting is due to the removal of mental content from consciousness because it causes too much anxiety when attention is focused on it. In other words, this theory states that a person may transform an explicit memory into an implicit one when the information contained in the memory distresses him or her. According to this theory, people do this because, if they no longer are aware of the memory, then their distress disappears. For example, if you had seen one of your parents temporarily become very ill when you were a young child, your memory for this event (even after the parent had recovered) may have caused you a great deal of anxiety. If you were not able to cope at this time with your memory of the event (for example, if you were too young to be able to say to yourself something like, “well, that was upsetting, but now it is over and everything is alright”), then you would have felt severe anxiety whenever you remembered it. At this point, according to defensive theory, you may have transformed the explicit memory into an implicit one by using what I will call “defensive forgetting.” If you do not now remember certain traumatic events from your childhood that you know happened (because others have told you they did), defensive theory states that defensive forgetting may be the cause.

Although the concept of repression is widely accepted among therapists and the general public, the evidence for it is not very good. There is one major problem with most research on repression: because repression is thought to work unconsciously — and, thus, has no unambiguous (clear) effects on our conscious mental events and behaviors — it is difficult to find examples of this defense mechanism that cannot be explained in other ways. One needs controlled research to rule out the various possibilities. As I stated earlier, the evidence supporting repression consists mostly of clinical case studies that use some version of recovered-memory therapy. Case studies, however, involve the use of uncontrolled research situations. The results of few (if any) laboratory studies using adequate controls give support to the concept of repression. As Holmes (1990) concluded in his review of decades of research on repression: “despite over sixty years of research involving numerous approaches by many thoughtful and clever investigators, at the present time there is no controlled laboratory evidence supporting the concept of repression” (p. 96). He stated that clinicians often dismiss these studies and instead focus on evidence involving case studies. Holmes, however, correctly argued that case studies do not provide adequate evidence for any claim, including claims about the existence of repression. He suggested in jest that anyone using the concept of repression should also provide the following cautionary label: “Warning. The concept of repression has not been validated with experimental research and its use may be hazardous to the accurate interpretation of clinical behavior” (p. 97).

Because of this lack of controlled experimental evidence for repression, cognitive psychologists have questioned whether or not repression actually occurs. They are especially concerned about those "memories" that return after weeks, months, or years of therapeutic work. They base their criticism on the evidence for reconstruction theory showing that, when we retrieve a long-term memory, we reconstruct it by combining the fragments of information contained in the activated engram with the large amount of information contained in activated schemas. In this view, each reconstructed memory is always a combination of fact and fiction. In other words, according to reconstruction theory, all of our memories are inaccurate to various degrees. When a person is strongly encouraged to retrieve memories of events that may not have occurred (as they are in recovered-memory therapy), it seems possible that memory reconstruction could lead to predominantly false memories — memories that include true but irrelevant details surrounded by a fictional story. For example, clients receiving recovered-memory therapy often are encouraged to visualize traumatic childhood events that they suspect may have occurred. It is plausible that such a technique could cause the person to incorporate imagined events into the reconstructed “memory.” Ofshe and Watters (1994) suggested that the process could occur something like this:

Picture an elephant. Imagine an apple. Now spend a moment visualizing an image of being sexually assaulted by one of your parents. It is an often distressing trick of the mind that it will create any event regardless of our desire to visualize that event. What separates an imagined image from a memory image is not a simple matter, for even imagined events are themselves largely built from memory. Our ability to imagine an elephant would be impossible if we didn’t have a memory of having seen an elephant or a picture of the creature. Similarly, our ability to ... imagine a sexual assault by a parent would also come from an amalgam of memories. To create this image we might use recollections of our parents’ physical appearances and of ourselves as children. We might place the scene in the memory of our childhood room. To create the action of the scene, we might use memories of other people’s descriptions of sexual assaults or of abuse scenes depicted in books or movie dramas. In the end, all the pieces of the imagined event would have something of the weight of memory. (p. 107)

These imagined events would be combined with bits of remembered events from childhood (for example, the night one of your parents came into your bedroom to open your windows after the air conditioner broke) to produce a remembrance that contains a small slice of truth and a large amount of fiction — a remembrance that accords with the therapist’s suggestion that you have repressed memories of traumatic events.

Study Questions

44. What is the cause of forgetting according to reconstruction theory?
45. What is an example from your own life of forgetting because of reconstruction?
46. What is the best way to prevent forgetting due to the reconstruction of memory codes?
47. How would you define the concept of "flashbulb memory" in your own words?
48. What is an example of a flashbulb memory not mentioned in the text (perhaps one that you've experienced)?
49. What did Neisser and Harsch (1992) study and how did they study it?
50. What were the results of the study by Neisser and Harsch (1992) and how did they interpret these results?
51. What is the cause of forgetting according to encoding-specificity theory?
52. What is an example from your own life of forgetting because of a mismatch between a retrieval cue and the way in which a memory was encoded?
53. What is the best way to form a memory code that can be activated by a broad range of retireval cues?
54. What is the cause of forgetting according to interference theory?
55. What is an example from your own life of forgetting something due to interference?
56. What is the best way to prevent forgetting caused by confusing similar information stored in explicit memory?
57. What is the cause of forgetting according to defensive theory?
58. What is an example of defensive forgetting that is not mentioned in the text?
59. Why is it so difficult to find clear evidence for defensive forgetting?