MEMORY

Overview of the Memory System

The multi-store model of memory developed by Richard Atkinson and Richard Shiffrin has guided research in memory and its development. The model is supported by extensive experimental evidence and is applied productively in work with individuals who have suffered brain injuries and students with typical learning characteristics as well as learning difficulties. In this information-processing model, illustrated in Figure 1, human memory is seen as operating in a manner analogous to that of a computer. The model depicts three separate memory stores that function as the hardware of the memory system: long-term memory, the sensory register, and working memory. Long-term memory, which is what people typically mean when they refer to memory, is a relatively permanent memory store with an apparently limitless capacity. It includes both semantic memory, a mental reference book that contains facts about the world, and episodic memory, a repository of stored traces of experienced events. It should be noted, however, that representations of learned material or personal experiences do not enter long-term memory directly. Information is moved through earlier stores to long-term memory.

Sights and sounds from the world enter the memory system through the sensory register. This store holds the icon of a visual display or the echo of a sound for a very brief period of time. Within only one second, information that an individual has not extracted from the icon or echo is lost. Hence, much of what enters the senses never becomes part of long-term memory. Information that is identified by individuals is often maintained in working memory while the individual interprets it, transforms it, or uses it to solve problems. Applying the computer metaphor, the contents of working memory correspond to open files and running programs. Information moves from working memory into long-term storage when it is processed in a way that makes it meaningful to the individual.

The long-term memory store is often incorrectly described as containing complete and actual representations of past events. In contrast, long-term memory is understood by psychologists to be a reconstructive process. Memories can be altered when individuals encounter new material that interferes with stored information or make inferences that are added to a representation. A well-documented way in which inference occurs is through the operation of schemas, which are organized sets of facts (for example, beliefs about what happens during a visit to the doctor). In a classic study conducted in 1932, Frederick Bartlett read English research participants a story that described events that were inconsistent with their own life experiences and hence seemed bizarre. When they were later asked to recall the story, the participants distorted the actions that took place in a manner that made them consistent with their own culture.

In addition to the memory stores described above, mental strategies are an important component of the information processing system. Strategies correspond to the software of the computer. They are learned procedures that individuals use to direct attention, move information from working to long-term memory, or apply techniques for solving problems. A second grader who repeats a telephone number before dialing it is using verbal rehearsal, a simple memory strategy; a ninth grader who learns a new vocabulary word by using it to describe a principle he has previously learned is using elaboration, an especially effective strategy. Strategies make it possible for memory to be intentional, the term applied when information is deliberately learned or purposefully retrieved. In contrast, memory is described as incidental when information is acquired in the process of performing another activity.

It should be apparent from even this brief introduction that the components of information processing are constantly interacting as a system. The contents of the long-term store are important in encoding, the process of taking information from the world into the memory system. Information stored in long-term memory makes it possible to identify the stimuli in the sensory register so that it can be moved to another store. Knowledge from long-term memory enables the application of strategies or problem-solving procedures in working memory. The contents of permanent memory also determine to a large extent how meaningful new material is, and hence how it can be organized and accessed from the long-term store.

The information-processing model is particularly useful in understanding explicit memory, the type of memory of greatest interest to parents and teachers. Explicit memories are potentially conscious and can be described verbally or pictured in images. A child uses explicit memory when she describes a class field trip or edits a report by applying grammatical rules she has learned. In contrast, some information may affect task performance without entering the individual's conscious awareness. In this case, it is described as implicit memory. Implicit memory is assessed by indirect measurements, such as determining how much quicker individuals can add letters to complete a word when they have previously been presented with a target word, or by physiological indicators, such as changes in galvanic skin response.

Understanding children's memory requires identifying the age-related changes that occur in the components of the information processing system. The first significant advance, of course, is the emergence of memory in development. In the next section, the point in life at which the memory system can be considered to be "up and running" will be discussed. The following sections will examine the changes that occur during childhood in the memory stores and in the use of strategies. The importance of these transitions in children's everyday lives will be explored in each section.

The Developmental Onset of Memory

The memory stores described above are assumed to be universal and present throughout life. Indeed, there is evidence that the capacity to store information in long-term memory begins even before birth. In a well-controlled investigation conducted by Anthony DeCasper and Melanie Spence, the researchers asked pregnant women to read aloud a Dr. Seuss story during the last six weeks of their pregnancies, a point in prenatal development at which fetuses can hear. Shortly after birth, the newborns' recognition memory was tested by comparing their reactions to the familiar passages versus similar but new story excerpts, both of which were read by the babies' mothers. The assessment built on the knowledge that babies can learn to modify the time between bursts of sucking when a change in sucking is followed by the presentation of a stimulus that serves as a reinforcer. The newborns wore headphones and were given a pacifier that recorded their sucking bursts. They indeed modified the way they sucked when the change in sucking was followed by the familiar passage, but they did not do so with the unfamiliar passage. The fact that the old, but not the new, passages served as reinforcers demonstrated that the babies could recognize the stimuli to which they were exposed before birth.

Although even very young infants can recognize sights, sounds, and smells they have previously encountered, the ability to recall an object or an experience develops later. Recall differs from recognition in that it requires coming up with a response as well as determining that it is correct. Some simple recall is present in the second half of the first year. As every babysitter knows, very young infants remain calm when their parents go out; by around seven months of age, however, separation protest is apparent. By about nine months of age, babies can imitate an action after a twenty-four-hour delay. Note that early recall is heavily dependent on cues and is limited to relatively brief time intervals. Recall continues to develop over the second year of life, corresponding to the development of the prefrontal cortex and other brain structures associated with explicit memory. Between age two and two and a half, toddlers can be expected to remember to stay away from common hazards, provide their first and last names when asked, repeat parts of nursery rhymes, and possess simple event schemas (also called scripts) for everyday events.

By age two and a half, as is well documented in the work of Robyn Fivush and her colleagues, children describe specific past experiences such as a trip to an amusement park. Such early memories, however, do not generally become a permanent part of autobiographical memory, the subset of episodic memory that represents individuals' own life histories. Most people do not recall anything that happened before they were three years of age. This phenomenon is described as infantile amnesia. Although the reasons for infantile amnesia are not completely understood, several factors appear to be important in explaining the developmental emergence of autobiographical memory. One of these is the maturation of the frontal lobes of the brain, which continues throughout early childhood. A second factor appears to be the emergence of the self-concept, which serves as a conceptual framework for the organization of memories. Another is the role of social interactions in maintaining early memories. Katherine Nelson has emphasized the importance of the child's participation in family discussions about past events for keeping early memories alive, a process described as reinstatement. Children are about three years old before they can actively participate in conversations about past events. Finally, some early memories may not be retrieved at later points in development because they are not effectively cued. Because the typical everyday environments of older individuals differ greatly from those of very young children, some potential memories may not be accessed.

Developmental Changes in Information Processing Capacity

Does the absolute capacity of the memory stores change with development? The answer to this question is unknown and, according to some experts, may be unknowable. It is clear, however, that memory span, a measure of working memory, improves reliably with age. Memory span is defined as the number of single words or digits individuals can report in order immediately after hearing them. It increases from about two items at age two to five items at age seven to seven items in adulthood. The amount of information that individuals can hold in memory at any one time determines at least in part what they can learn.

It appears that age-related improvements in the memory span are attributable to two factors: changes in the resources that are available in the information processing system and differences in the efficiency with which older children and adults apply these resources. One of the most important developmental changes in the information processing system is in the speed of processing. Robert Kail asked individuals from ages six to twenty-one to perform a variety of cognitive tasks ranging from mental addition to memory search. He found a remarkably similar pattern of age-related increases in reaction time in the performance of all of the tasks. It should be noted that speed of processing is important in executing many cognitive functions, from moving information from the sensory register to working memory to applying strategies. Hence, developmental differences in speed of processing can influence children's capabilities to perform a variety of academic and everyday tasks.

Processing capacity increases if more resources are available; it can also be enhanced if the resources necessary to perform the task can be decreased. Simply providing extra time for study or problem solving improves children's performance in some but not all tasks. Perhaps the most important determinant of task demands is the learner's relevant knowledge. Familiarity with the to-be-remembered material affects the memory process at every step of information processing. Imagine how difficult it would be, for example, for someone who speaks only English to remember a string of digits presented in another language.

Michelene Chi provided a classic demonstration of the importance of prior knowledge. In this investigation, children who were chess experts had better memories for chess positions than did nonexpert adults, even though the adults demonstrated the expected advantage in remembering digits. Here, prior knowledge of chess eliminated age differences in memory. Because children's knowledge bases increase as they get older simply through their experiences in the world, age differences in memory may be explained to a great extent by differences in task-relevant knowledge.

Changes in the knowledge base also affect the organization of information in long-term memory and hence partially determine whether or not material can be reported. As depicted in the semantic network model of memory, items are represented as nodes and the nodes for associated items are linked. Knowledge about a particular content domain (such as dinosaurs or soccer) creates semantic associations between previously unconnected items, resulting in the formation of connections between nodes. When one item is recalled, the activation spreads to associated items as well, increasing the likelihood that they will also be reported. Items that are represented as part of an extensive network of information within a content area are also less likely to be lost over time. Moreover, age differences in children's long-term event recall are minimized when younger and older children have high levels of prior knowledge about the to-be-remembered components of the experience.

Knowledge also enhances memory by making it easy to apply strategies for encoding and retrieving information. As discussed in the next section, the development of memory strategies is another important reason for age-related increases in children's ability to remember, especially in academic settings.

The Acquisition of Memory Strategies

A child's intentional memory shows dramatic improvement when he can effectively use memory strategies. These deliberate tactics for remembering develop over a lengthy period that spans the elementary and middle school years. Preschool children use very simple tactics for remembering in some special task settings; for example, a four-year-old can be expected to use a marker to denote an object's hiding place in preparation for subsequently finding it. Pre-schoolers do not, however, use mental strategies and indeed do not typically differentiate memory and perception. By age seven, most children spontaneously use rehearsal to enhance short-term memory performance. Retrieval strategies (such as going sequentially through the alphabet) begin to be spontaneously used around third grade. Children's self-directed use of organization, the ability to impose a semantic structure on the to-be-remembered items to guide memory performance, emerges later in the elementary school years. For example, a fifth grader could be expected to remember what to take on a school field trip by reviewing the activities involved and packing accordingly (e.g., "things to do on the bus," "what I need for the nature walk," "things I'll need at lunch") by herself. It is unlikely, however, that a third grader would use organization as a guide without the direction of an adult. Elaboration, a highly effective strategy that involves actively creating a representation integrating new information with existing information, may not be used spontaneously by children until early adolescence. An eighth grader may remember the definition of the word "fruition" by creating an image of a ripe pear on a bough, but it is unlikely that most sixth graders would do so.

Most students do not acquire optimally effective study skills, which can be considered forms of memory strategies, until adolescence. By the later elementary school years, children allocate more study time to test items that they have previously missed on practice tests; in contrast, younger elementary school children devote comparable amounts of attention to items that they have previously gotten correct and incorrect. By high school good students can be expected to take spontaneous and effective notes and to emphasize key information in studying.

Regardless of the strategy under consideration, its use emerges initially in highly supportive task settings. At early points in the acquisition of memory strategies, children use a strategy when adults instruct them to use the strategy but they do not generate the strategies spontaneously, a limitation described as a production deficiency. Additional preparation time, the use of familiar materials, and cues that may reduce memory demands are other factors that increase the likelihood that strategies will be used. When they are fully acquired, strategies are applied spontaneously in a wide array of task settings. Metamemory, or knowledge about the operation of the memory system, contributes to the development of strategies. Metamemory involves understanding the demands of the task and the relevant characteristics of the rememberer, as well as identifying and using effective task approaches and monitoring their effectiveness. In general, the development of metamemory increases with age, corresponding to changes in actual memory performance.

Fostering Children's Memory Development

Memory develops largely through social interaction, and consequently parents and teachers play a critical role in assisting children in developing good memories. During early childhood, everyday adult-child interactions provide the basis for memory development by fostering language development and helping children acquire scripts for common events. Conversations that include reminiscing about the recent past are significant in helping children understand what memory is and what should be remembered. Developing family traditions and participating in rituals encourage reminiscing.

When a child begins formal schooling, adults can contribute to the child's acquisition of metamemory and memory strategies by modeling their own approaches to tasks involving memory and discussing their memory failures. It is important that teachers and parents recognize that strategies develop over an extended period. Children's use of effective memory strategies in studying should be monitored, and adults should help children simplify tasks by providing external memory aids and other supports as needed. Informal instruction in the use of memory strategies should be incorporated into class presentations and assignments.

It is also important to keep in mind that memory is facilitated by advances in other domains of development. As examples, narrative skills increase children's abilities to provide reports of their personal experience; problem-solving techniques increase functional working memory capacity. The development of the knowledge base plays a vital role in memory performance. Hence, providing children with opportunities to learn about the world contributes significantly to their capacity to remember effectively.

See also: COGNITIVE DEVELOPMENT; LEARNING

Bibliography

Atkinson, Richard C., and Richard M. Shriffrin. "Human Memory:A Proposed System and Its Control Processes." In K. W. Spence and J. T. Spence eds., The Psychology of Learning and Motivation: Advances in Research and Theory. New York: Academic Press, 1968.

Chi, Michelene. "Knowledge Structures and Memory Development." In R. S. Siegler ed., Children's Thinking: What Develops? Hillsdale, NJ: Lawrence Erlbaum, 1978.

DeCasper, Anthony J., and Melanie J. Spence. "Prenatal Maternal Speech Influences Newborns' Perception of Speech Sounds. Infant Behavior and Development 9 (1986):133-150.

Fivush, Robyn, and Judith A. Hudson, eds. Knowing and Remembering in Young Children. Cambridge, Eng.: Cambridge University Press, 1990.

Howe, Mark. The Fate of Early Memories: Developmental Science and the Retention of Childhood Experiences. Washington, DC: American Psychological Association, 2000.

Kail, Robert. "Development of Processing Speed in Childhood and Adolescence." In H. W. Reese ed., Advances in Child Development and Behavior. San Diego: Academic Press, 1991.

Nelson, Katherine. "The Psychological and Social Origins of Autobiographical Memory." Psychological Science 1 (1993):1-8.

Molly Carter Imhoff

Lynne E. Baker-Ward