What is long-term memory?

Introduction

William James, in his famous work
The Principles of Psychology
(1890), was one of the first to make the distinction between short-term memory and long-term memory, which he called primary and secondary memory. James believed that secondary (long-term) memory was the only true memory, because it possessed two important characteristics that are absent from primary memory: remembered events seem to belong to the past, and their recollection is brought about by appropriate cues. Another early memory researcher was Hermann Ebbinghaus, who introduced the use of nonsense syllables (meaningless sets of two consonants and a vowel) in investigating the nature of long-term memory. Ebbinghaus systematically studied forgetting of information from long-term memory and found that most forgetting occurs during the first nine hours after learning, and especially during the first hour. After that, forgetting continues, but at a much slower rate. He also discovered that much forgetting from long-term memory is caused by interference from other, previously learned material (proactive interference) or by interference created by learning new material (retroactive interference).

Psychologists today generally propose a three-stage theory of memory: sensory memory, short-term memory, and long-term memory. When information first enters through one of the senses, it is retained briefly (for less than a second, generally) in sensory memory. Even though this information fades very rapidly, through processes such as selective attention, some of the information is processed further. The next processing stage is referred to as short-term, or working, memory. Information is retained in this stage for about twenty to thirty seconds, but through the use of rehearsal, items can be maintained in short-term memory indefinitely. Short-term memory has a limited capacity; it can hold about five to nine items at a time.

Information can reach long-term memory by several methods. Items that are particularly meaningful or that have a high emotional content are usually directly encoded into long-term memory. The use of a type of rehearsal called elaborative rehearsal, which involves thinking about how new material relates to information already stored, is also an effective method for transferring items from short-term memory to long-term memory. Unlike short-term memory, storage in long-term memory appears to be relatively permanent, and the capacity of long-term memory appears to be virtually unlimited.

In 1972, Fergus Craik and Robert Lockhart suggested that long- and short-term memory do not necessarily represent distinct stages of memory. They proposed what has come to be known as the levels-of-processing model, which holds that differences in how long or how well something is remembered depend on the degree or depth to which incoming information is mentally processed. The depth of processing for incoming information is related to how much it is thought about, organized, and related to one’s existing knowledge. Thus, long-term memory, in their view, simply represents information that has been processed to a greater depth. This model has been very influential and has stimulated a tremendous body of research.

Classifying Long-Term Memory

Another influential theorist, Endel Tulving, proposed a classification scheme that distinguishes three aspects of long-term memory: procedural memory,
episodic memory, and semantic memory. Procedural memory, also called skill memory, represents knowledge of how to do something. This can involve motor skills, such as knowing how to swing, or cognitive skills, such as reading and writing. Episodic memory is the memory of events from one’s personal past. One’s recollection of where one was last Saturday night or what one had for dinner yesterday would represent information from episodic memory. Semantic memory represents a person’s knowledge of the world not tied to a specific event in one’s life. This type of general knowledge includes definitions of words, facts such as the name of the first president of the United States, and relationships between concepts. Although some researchers have questioned the need for these distinctions among types of memory, there is evidence that supports this classification scheme. Many people who suffer from amnesia (partial or total loss of memory), for example, usually are unable to remember specific incidents or facts about their lives (episodic memory), yet their general knowledge (semantic memory) and knowledge of how to do things (procedural memory) remain intact. Also, studies involving monitoring of blood-flow patterns in the brain reveal different patterns when one is thinking about personal experiences versus impersonal facts. The frontal region of the brain appears to be more active during retrieval from episodic memory, and the posterior regions have a greater degree of activation during semantic retrieval.

Encoding, Storage, and Retrieval

Researchers have also attempted to understand the nature of three basic processes associated with long-term memory: encoding, storage, and retrieval. Encoding concerns how information is put into memory; storage refers to the maintenance of information in memory; retrieval refers to the recovery of information from memory. Forgetting can represent a failure of any of these processes. Researchers are interested in the specific mental operations that are involved in each of these operations. Successful encoding of information into long-term memory is usually based on the meanings of the items. This is in contrast to short-term memory, which primarily involves an acoustic coding of information. Acoustic codes can also be used in long-term memory, but the dominant or preferred code involves meaning.

As mentioned above, the capacity of long-term memory appears to be virtually unlimited, and storage relatively permanent, although some storage loss probably does occur. Experimental evidence indicates that storage processes usually require some time for items to be placed in long-term memory. This process is referred to as consolidation. If a disruption (such as electroconvulsive shock) occurs shortly after information is encoded, the information will be lost from storage. Sometimes, information is definitely stored in one’s long-term memory but one still cannot recall it (for example, the name of an old friend whom one has not seen for some time). Usually, this represents a retrieval failure. Retrieval of information from long-term memory has sometimes been compared to searching for an item stored in an attic. The success of the search process depends on knowing where to look. The retrieval of information stored in long-term memory is usually initiated by some stimulus, referred to as a retrieval cue. Such cues provide information that can aid in the recall of stored material. Factors associated with encoding can also aid in retrieval. For example, organizing information at the time of encoding will increase the chances of successful retrieval. Retrieval will also be facilitated if the context in which information is encoded is similar to that in which it will be retrieved. Many studies have even shown that the psychological state (for example, the drug state or mood state) that one is in during the time information is encoded can act as a retrieval cue at the time of recall. This is referred to as state-dependent memory. Thus, if one was depressed when one learned material, one will more easily recall that material when one is in that same state again.

Eyewitness Testimony Research

Investigations of the nature of long-term memory have led to findings that have proved useful in many areas of psychology and in other fields. Elizabeth F. Loftus and other researchers have done extensive research on the accuracy of long-term memory in eyewitness testimony. Loftus found that juries tend to rely heavily on eyewitness testimony, but she also found that the reliability of the memories of eyewitnesses was not very high.

When episodic information is stored in long-term memory, it is constantly being affected by material stored in semantic memory, including beliefs, prejudices, expectancies, inferences, and so on. What is stored in memory is not a simple copy of experiences but rather a construction that a person has created based on material already stored in long-term memory. People frequently fill in gaps in their memory with appropriate bits and pieces of their general knowledge of human situations and activities. These changes occur without their awareness; a memory of an event might seem accurate yet may have been changed dramatically. Loftus showed that the memory of witnesses could be affected by the type of question asked or by the specific wording of a question. In one experiment, subjects were shown a film of a traffic accident and then were asked questions about it. A question about the speed of the vehicles was asked in two different ways: “How fast were the cars going when they smashed into each other?” and “How fast were the cars going when they hit each other?” Subjects who were asked the “smashed” question not only estimated greater speeds but also one week later had memories of many details (such as broken glass) that were not in the film.

Another study, done on television station WNBC in New York, allowed viewers to witness a simulated crime in which they could closely see the assailant’s face for 3.5 seconds. They were then asked to pick out the criminal from a lineup of six men through a special telephone number. Eighty-six percent of the 2,145 viewers who phoned in either “recognized” the wrong man or decided that the guilty man was not in the lineup. Overall, the performance was at the level that would be expected by chance alone. This type of research has led to modifications in some courtrooms, such as clearer instructions to jurors that are more easily remembered (jurors are usually not allowed to take notes, so they have to rely wholly on long-term memory).

Mnemonic Techniques

Understanding the nature of long-term memory has led to the development of many mnemonic techniques: strategies to help improve memory. Two mnemonic systems that are particularly useful for remembering ordered sequences of unrelated items (such as grocery lists) are the method of loci (loci is Latin for “places”) and the peg-word system. In the method of loci, the first step is to think about a set of familiar geographic locations, such as the rooms in one’s house. The next step is to form a mental image of each item on the list in one of these locations, that is, one word in each room in the house. Whenever one wants to remember the list, one takes a mental walk through the house and collects the items in each room.

Use of the peg-word system requires the memorization of a list of words that will serve as memory pegs for the list to be remembered. One popular example of a peg-word list is: “One is a bun, two is a shoe, three is a tree, four is a door, five is a hive, six is a stick, seven is heaven, eight is a gate, nine is a hive, and ten is a hen.” Once this list is learned, mental images are created between each new item to be remembered and the previously learned peg word. For example, if the first item on a grocery list is coffee, one might image coffee being poured over a bun. If the second item is milk, one might image a milk carton being kicked by a shoe. This might seem cumbersome, but it works. With both the method of loci and the peg-word system, particularly vivid images seem to produce the best results.

Learning Techniques

The study of long-term memory has also led to the development of techniques for improving the learning of more complex material. In general, the more that items are elaborated on during the encoding, the more easily they can subsequently be recalled or recognized. This is true because the more connections are established between items, the larger the number of retrieval possibilities. Therefore, if one wants to remember some fact, one should expand on its meaning. Questions about the causes and consequences of an event, for example, are particularly effective elaborations because each question sets up a meaningful connection, or retrieval path, to the event.

The more organized the material is during the process of encoding, the more easily the material will be retrieved later. Massive amounts of information can be stored and retrieved, if only it is properly organized. For example, remembering textbook material can be facilitated by not only outlining a chapter but also sketching a hierarchical tree that pictures the relations between chapter headings and subheadings.

A formal strategy developed for remembering textbook material is the SQ3R method. SQ3R stands for “survey, question, read, recite, and review.” Before reading a chapter, one should survey, or skim, the material, looking for section headings and key ideas. Next, one should question or ask oneself what subject will be covered before one reads each section. As one reads the text, one should continue thinking about the question and how topics are connected. Next, one recites to oneself the major points at the end of each section. Finally, one reviews all the material after one has read to the end of the chapter. This method encompasses most of the principles by which material can be effectively encoded, stored, and retrieved from long-term memory.

Evolution of Study

The study of long-term memory has continued in psychology since the pioneering work of James and Ebbinghaus in the late nineteenth century. It was clear even during those early efforts that a distinction could be made between the temporary storage of information in short-term, or working, memory, and the relatively permanent nature of long-term memory.

During the early part of the twentieth century, however, the study of memory fell into disfavor during the era in which psychology was dominated by the behaviorist movement. Behaviorists believed that if psychology was to be a science, it had to limit its investigations to phenomena that are directly observable and measurable. Memory was viewed as a subjective element that was not amenable to experimental investigation.

It was not until the development of the computer, around the time of World War II, that the experimental study of memory again achieved respectability in academic psychology. The computer offered an objective model of how information might be stored and processed in human memory. Since that time, developments in computer science and in the psychological study of memory have progressed in tandem, with breakthroughs in one area providing insights into the other. The computer model is still very much a part of the study of memory, with new theories of memory often modeled as computer programs. In many ways, the study of memory was responsible for experimental psychology shifting from a focus exclusively on overt behavior to the “cognitive revolution,” in which mental functions are studied directly.

The study of long-term memory in the future will likely focus more on understanding how processes related to memory have their physiological basis in the functioning of the brain. Locating where memories are stored in the brain has long eluded psychologists. Most now believe that memory is distributed over the entire surface of the cerebral cortex rather than located at a particular spot. It is also known from the study of amnesiacs that brain structures such as the hippocampus and amygdala appear to be involved in the consolidation of new material in long-term memory. As knowledge of neurotransmitters (chemicals involved in the sending of messages from one brain cell to another) expands, the role that they play in memory may also become more apparent.

Bibliography

Baddeley, Alan D. Your Memory: A User’s Guide. London: Carlton, 2004. Print.

Bransford, John. Human Cognition: Learning, Understanding, and Remembering. Belmont: Wadsworth, 1989. Print.

Fedoseev, Lazar M., and Arseni K. Alexandrov. Long-Term Memory: Mechanisms, Types and Disorders. New York: Nova Science, 2012. Digital file.

Höchsler, Christian, ed. Neuronal Mechanisms of Memory Formation: Concepts of Long-Term Potentiation and Beyond. New York: Cambridge UP, 2005. Print.

Loftus, Elizabeth F., et al. Eyewitness Testimony: Civil and Criminal. 5th ed. New Providence: LexisNexis, 2013. Print.

Neath, Ian. Human Memory: An Introduction to Research, Data, and Theory. Belmont: Wadsworth, 2006. Print.

Neisser, Ulric, and Ira E. Hyman, eds. Memory Observed: Remembering in Natural Contexts. 2d ed. New York: Worth, 2000. Print.

Parkin, Alan J. Memory and Amnesia: An Introduction. N.p.: Taylor, 2013. Digital file.

Schweppe, Judith, and Ralf Rummer. "Attention, Working Memory, and Long-Term Memory in Multimedia Learning: An Integrated Perspective Based on Process Models of Working Memory." Educational Psychology Review 26.2 (2014): 285–306. Print.

Vitevitch, Michael S., et al. "Complex Network Structure Influences Processing in Long-Term and Short-Term Memory." Journal of Memory and Language 67.1 (2012): 30–44. Print.