Memory Structures and Processes

Cognitive Psychology: Theory, Process, and Methodology - Dawn M. McBride, J. Cooper Cutting 2019

Memory Structures and Processes

Questions to Consider

· Is memory a process, a structure, or a system?

· How many types of memories are there?

· Are there differences in the ways we store and retrieve memories based on how old the memories are?

· What kind of memory helps us focus on a task?

· How does our memory influence us unintentionally?

· What are the limits of our memory?

Introduction: The Pervasiveness of Memory

Memory is pervasive. It is important for so many things we do in our everyday lives that it is difficult to think of something humans do that doesn’t involve memory. To better understand its importance, imagine trying to do your everyday tasks without memory. When you first wake up in the morning you know whether you need to jump out of bed and hurry to get ready to leave or whether you can lounge in bed for a while because you remember what you have to do that day and what time your first task of the day begins. Without memory, you would not know what you needed to do that day. In fact, you would not know who you are, where you are, or what you are supposed to be doing at any given moment. It would be like waking up disoriented every minute.

There are, in fact, individuals who must live their lives without the aid of these kinds of memories. An extreme case is the story of Clive Wearing, a man in the United Kingdom who suffered a brain injury due to an illness from encephalitis (see Photo 5.1). From the illness, the area of his brain known as the hippocampus and the surrounding brain tissue were damaged. The hippocampus is a brain structure that is very important in storing and retrieving memories. Due to this damage, Clive lost the ability to know what was going on around him for more than about a minute at a time. He described his life as if he were just waking up every moment. He has to continuously figure out what is going on around him. Imagine having the experience of suddenly becoming consciously aware of yourself and your surroundings, but everyone else around you is acting normally and not paying any attention to your wakening. It is like waking up from being in a coma for many years and yet no one is standing around you explaining what has happened. You have to try to figure it out for yourself with no context or knowledge of what has occurred in the previous few moments. Imagine how frustrating this would be! Interestingly, Clive retains his musical abilities (e.g., playing the piano), at which he was an expert before his illness. This illustrates one of the important differences in types of memories: There are those about episodes in one’s life, known as episodic memories, and those about skills we have developed over time, known as procedural memories. We talk more about each of these types of memories in this and the next chapter, in addition to other types of memory such as memory for general knowledge and facts (semantic memory), memory about one’s self (autobiographical memory), and memory for tasks we intend to perform in the future (prospective memory). Memory deficits, including amnesia, are also discussed further in Chapter 6.

Memory as Structure or Process

Memory can be thought of in many different ways. As described in Chapter 1, Aristotle thought of memory as similar to a wax tablet that can be molded, melted, and remolded over time. Memory can also be thought of as a filing system for information organized in different ways (e.g., all the animals are stored together, all the colors are stored together), depending on how it is encoded and how it is retrieved. Both of these ideas view memory as a “thing,” as a storage unit or structure where information is held. However, memory can also be thought of as a collection of interdependent processes. In other words, rather than thinking of memory as a thing, memory is thought of more as “remembering,” and researchers who adhere to this view of memory focus more on how and when remembering occurs, rather than memory as a storage structure or unit. Structure and process views of memory have both been important in how researchers have studied memory, and you will see examples of both views as we consider some of the types of memory researchers have investigated in this and the next chapter.

Photo 5.1 Clive Wearing and his wife, Deborah


Ros Drinkwater/Alamy Stock Photo

Encoding, Storage, and Retrieval

Three important processes in memory are encoding, storage, and retrieval. Encoding is the process by which information enters our memory. It is sometimes a fairly active process, such as you reading this text or quizzing yourself to try to remember the material you are leaning in a course. It can also be a less active process when information is encoded without one intending to remember it. However, in order for information to be encoded, attention to the information is often required (see Chapter 4 for descriptions of attention processes). Storage is the process by which information is kept in memory. Connecting with one’s preexisting knowledge seems to be important in the retrieval process, as information seems to be stored with related concepts (see Figure 5.1). However, there is no single place in the brain where an individual memory is kept. Instead, the storage of memories seems to be distributed across multiple brain areas. Specific brain areas (e.g., the hippocampus) are involved in pulling the pieces of a memory back together when it is retrieved (see Photo 5.2). Like encoding, the retrieval process can be intentional, such as when you attempt to remember what you had for breakfast last Thursday or the name of the instructor of your course, or unintentional, such as when you suddenly remember the correct answer to an exam question at 3:00 a.m. the day after you took your exam. Figure 5.1 summarizes the processes of encoding, storage, and retrieval.

Encoding: the process of inputting information into memory

Storage: the process of storing information in memory

Retrieval: the process of outputting information from memory

Photo 5.2 Hippocampal activity in the human brain during retrieval.


Reprinted with permission from “Conscious Recollection and the Human Hippocampal Formation: Evidence From Positron Emission Tomography,” by D. L. Schacter et al., Proceedings of the National Academy of Sciences, USA, 93, 321—325. Copyright 1996 National Academy of Sciences, USA.

Figure 5.1 Encoding, Storage, and Retrieval


Source: Photo from BananaStock/BananaStock/Thinkstock

Cognitive neuroscience studies show that these three processes are controlled by different brain areas (Moscovitch, Chein, Talmi, & Cohn, 2007). Let’s consider the processes of encoding, storing, and retrieving for a scene from my home office that is occurring as I type: my dog, Daphne, lying in her bed. Encoding of the visual information of the scene takes place initially in the primary visual cortex in the occipital lobe (described in Chapter 3). From there, the visual information is processed in my medial temporal lobe, where the visual information binds to other sensory information from other areas of the cortex (e.g., the sound of her snoring as she sleeps, the concept knowledge I have about dogs). The binding of the information stored in these different cortical areas will aid in putting those pieces back together later when I want to remember the scene. When I turn back around to my computer and attempt to recall the scene of Daphne in my mind, the area of my visual cortex where the visual information of that memory is stored becomes active, along with the sound of her snoring stored in the temporal cortex that this information was bound to when it was stored. With help from the medial temporal lobe area (especially the hippocampus—see Photo 5.2), these different areas that were bound together when the elements were stored will each become reactivated to allow me to retrieve the encoded memory.

Modal Model of Memory

In addition to the descriptions of memory we have already discussed, memory has also been classified according to duration: sensory memories, short-term memories, and long-term memories that describe very brief memories, fairly brief memories, and longer-held memories, respectively. An early model of memory known as the modal model of memory (Atkinson & Shiffrin, 1968) describes these types of memories along with hypothetical structures that hold memories for different lengths of time. Figure 5.2 illustrates the modal model of memory with information coming in through our senses into sensory memory, being passed on to short-term memory when attention is given to the information, and finally being stored well in long-term memory if the information is processed in connection with other knowledge already stored there (elaborative processing). Each of these three types of memories is described in this chapter, along with the methods researchers have used to study them.

Figure 5.2 Atkinson and Shiffrin’s (1968) Modal Model of Memory


Sensory Memory

Sensory memory is the briefest form of memory. It includes memories of raw, unprocessed sensory information. If you focus your eyes on a bright scene (e.g., looking out the window) and then close your eyes, you will see a brief afterimage of the scene that fades very quickly (see Photo 5.3 for another example). This is a sensory memory. It is a visual representation of the scene that exists in its sensory form and is lost from memory within a second or two. Sensory memories can be stored for very brief periods of time for each of our senses, but these memories have been very difficult for researchers to measure because of their brief duration. These memories are short enough that subjects in research studies typically do not have time to report their retrieval from sensory memory before the memory has disappeared. How then do we know about the capacity and duration of these memories?

Sensory memory: the very short-term memory storage of unprocessed sensory

Partial-report method: a research procedure where subjects are asked to report only a portion of the information presented

Photo 5.3 Sensory memory is like the trail of light that comes from shaking a sparkler around on a dark night.


Roger Ressmeyer/Corbis/VCG/Corbis Documentary/Getty Images

One of the first studies of visual sensory memory (also known as iconic memory) to help answer this question was conducted by George Sperling (1960). To allow subjects in his study to report enough of the memory to measure sensory memory capacity and duration, he asked them to report on a portion of what was presented to them. This method is known as the partial-report method because subjects are only asked for a partial report of what was presented. Figure 5.3 illustrates how the method was used in Sperling’s study. In this study, subjects were presented with arrays of letters for a very brief time (only 50 ms in one experiment) and then asked to report just one row of letters according to a tone (low for first row, medium for middle row, and high for top row). Based on how many letters subjects could report for that one row, he estimated how many they would have been able to report from the whole array if there had been enough time to do so before they faded from sensory memory. Thus, if subjects could report an average of three of the four letters in the row, then 75 percent (3/4) of the letters were available to them at the time they were asked to report them. With no delay between the end of the display and the instruction tone, subjects could remember and report an average of about 75 percent of the letters in the row they were asked to report. When asked to report all the letters (not just one row), subjects could only accurately report about four letters on average regardless of how many letters they were shown (e.g., if shown twelve letters, this is only about 33 percent of the letters in the whole array). With the partial-report method, Sperling showed that the capacity of the visual sensory memory is fairly large and much larger than had been measured previously. In subsequent experiments, Sperling systematically delayed the presentation of the tone to measure the duration of sensory memories. In these experiments, he learned that visual sensory memories are held for about one second. After this length of time, memory performance from a partial report declines to the level equal to the performance seen when subjects were asked to report the whole array (about four letters).

Stop and Think

· 5.1. Describe the three primary processes of memory.

· 5.2. List the three hypothetical storage structures of memory from the shortest to the longest storage.

· 5.3. Consider different ways in which you encode information you learn in class (e.g., visually, aurally). How effective do you think each of these encoding processes is for storing information in long-term memory?

After the Sperling (1960) study showed that visual sensory memories last about one second, other researchers examined the duration of sensory memory for nonvisual senses. For example, studies using the partial-report method that focused on auditory sensory memory (also known as echoic memory) reported that these memories could last as long as four seconds (e.g., Darwin, Turvey, & Crowder, 1974). Studies of tactile sensory memory (e.g., Sinclair & Burton, 1996) suggest that these memories last as long as five seconds. However, from the researcher’s perspective, it can be difficult to determine if subjects are reporting sensory memories involving unprocessed sensory stimulation or short-term memories that have been processed to some degree. In other words, where does sensory memory end and short-term memory begin? Because of this issue, it is unclear if the longer estimates for auditory and tactile sensory memory reflect sensory or short-term memories. In addition, the majority of research in sensory memory has focused on visual and auditory senses. Thus, little is known about sensory memory for the other senses.

Figure 5.3 Partial-Report Method of Visual Memory Studies


Sources: (a) Based on Sperling’s (1960) study design. (b) Photo from Digital Vision/Digital Vision/Thinkstock.

More recently, researchers have attempted to better understand how information is lost from sensory memory. One proposal is that there are two stages of sensory memory storage of different durations (Cowan, 1988). In the first stage, the raw, unprocessed perceptual information is stored, and in the second stage, the perceptual information connects with information stored in long-term memory that allows for interpretation of the stimuli. This description of sensory memory can explain the difference in results across the different sensory modalities: The duration of one second for visual sensory memory reported by Sperling (1960) represents the first stage of sensory memory, whereas the longer durations reported for auditory and tactile sensory memory represent the second stage of sensory memory.

Stop and Think

· 5.4. Explain how the partial-report method allows researchers to more accurately estimate the capacity of sensory memory than a whole-report method.

· 5.5. According to the research in this area, what is the duration of sensory memories?

· 5.6. Research in sensory memory for senses other than vision and audition is scarce. Imagine that you are researching olfactory (sense of smell) sensory memory to contribute to the gap in the research in this area. Describe a study you might design using the partial-report method to study olfactory sensory memory. What are some of the limitations of this method for this type of sensory memory?

Recent research in cognitive neuroscience has been providing new information about how sensory memory operates. For example, studies by Lu, Williamson, and Kaufman (1992a, 1992b) have shown that the decay of auditory sensory memory corresponds to decay in activity in specific areas of the brain responsible for processing auditory information (e.g., auditory cortex). Lu, Neuse, Madigan, and Dosher (2005) have also shown that visual sensory memories in individuals with mild cognitive impairments (such as those shown by individuals with early stage Alzheimer’s disease) decay faster than comparison individuals without these impairments. These studies suggest that there may be a link between the experience of a sensory memory and specific neural activity. Thus, research in sensory memory using methods from neuroscience is providing important new information about how these memories are formed and experienced and how to define a sensory memory.

Short-Term Memory (STM)

What were you just thinking about before you started reading this section? This memory is probably one stored in what is known as your short-term memory. Short-term memory (STM) is an intermediate memory storage that begins processing of perceptual information transferred from sensory memory. Information that becomes the focus of attention moves from sensory memory to STM. Clive Wearing, the amnesic described in the introductory section of this chapter, can hold memories in his STM for a short time, but once his attention moves on, those memories are lost. The term working memory is also used to describe the system that controls the processing and activation of the information held in STM (Nairne & Neath, 2013). We discuss the working-memory system later in this chapter because it was not a part of the original modal model of memory shown in Figure 5.2 and has its own model and research support.

Short-term memory: the short-term storage of memory with minimal processing that is forgotten quickly without elaborative processing

Information in STM can be held for a short time if it remains in the focus of attention (e.g., by rehearsing the information), but in order to store information for a longer period of time, the information must be transferred to long-term memory (e.g., by connecting the information to other information already stored in long-term memory). Processing of the information also affects the capacity of STM. When information is organized according to its meaning, more items can be stored in STM.

Consider this example: Look at the following numbers for a minute or so. Then close your eyes and try to recall them in order:

1 9 9 0 4 1 1 9 1 1 1 4 9 2 2 0 1 5

How many could you remember? Most people can remember about five to nine items stored in STM. Now, let’s try that again. This time when you look at the numbers, try to see if you can group them in some meaningful ways (e.g., as years or important numbers to call on your phone). Close your eyes and try to recall the numbers again. If you did not notice some meaningful organization the first time you studied them, you should have been able to increase your recall on the second try. In fact, if you were able to find important meaning in all of the numbers, you may have remembered all eighteen of them. This organizational processing likely more than doubled your initial recall level. The process of organizing information into fewer meaningful units is called chunking. You may have chunked the numbers together as 1990, 411, 911, 1492, 2015, leaving you with only five items to remember.

Chunking: a process of organizing information that allows more items to be stored in memory

Capacity of STM

This example illustrates the capacity of STM for most people: about five to nine items. This was famously shown by Miller (1956) in a study titled “The Magical Number Seven, Plus or Minus Two” that represents the average capacity of STM. Chunking works with other types of information as well. Letters can be grouped as words and words can be grouped as sentences to hold more items in STM. Miller measured STM capacity in a particular way. His seven-plus-or-minus-two number is based on the average number of items his subjects could recall accurately in the correct order 50 percent of the time. This is known as the span of STM and has been used by numerous researchers to measure the capacity of STM for different types of information. There are some limits on the span of STM based on the type of information being stored, however. For example, span is smaller for words with more syllables (e.g., hippopotamus) than for words with fewer syllables (e.g., horse; Simon, 1974). More recent research also suggests that STM span may be closer to three to five chunks in some cases, and that limits on our attention (i.e., information in our attentional focus at a given time) are linked to the number of chunks that can be successfully stored in STM (Cowan, 2001). Thus, the capacity of STM can depend on factors like the type of information and our attentional limits.

Duration of STM

In fact, our attention limits the duration of STM storage as well. Information enters STM when we focus our attention on specific information in our sensory memory. It disappears from STM when our attention moves on to the next thing we are thinking about. Thus, memories are held in STM for as long as our attention lasts. If we intentionally hold information in our focus of attention for a longer than usual period of time, we can increase how long that information stays in STM. This typically occurs through active rehearsal, which means repeating the information within our mind. This is represented by the curved arrow in Figure 5.2, showing that information can be recycled in STM through rehearsal. To illustrate this, suppose that you have stopped at the store to get a few items on a list you have stored on your phone. Your phone’s battery is dying so you take a quick glance at the list containing soda, chips, milk, bread, and cereal just before your phone’s battery dies. To remember the items as you go through the store, you may say the list to yourself (maybe just in your head, maybe not) over and over until you have all of the items in your basket. Then you can focus on paying for the groceries and retrieving the PIN of your ATM card. Once you focus on your payment, the list will likely be lost from STM, but because you have already gotten your items, the rehearsal has served its purpose.

Without rehearsal, the duration of STM is set by the typical time your attention stays focused on the information. But this attention can be given to information in degrees (as anyone who has worked on two tasks at once can attest). Thus, information is lost from STM gradually, rather than instantaneously. This was shown using a method originally developed by J. Brown (1958) and Peterson and Peterson (1959). In this method (see Figure 5.4), subjects are asked to remember a short sequence of letters, such as GRX. Meaningless strings of letters are used to prevent meaningful processing that might transfer the information to long-term memory. After hearing the letters, subjects are asked to complete a verbal interference task that typically involves counting down from a starting number (such as 576) by threes (e.g., 573, 570, 567). Counting is done for a variable amount of time to manipulate the delay time for recalling the letters. Peterson and Peterson (1959) had subjects count for three to eighteen seconds. A different string of letters was presented on each trial and then subjects counted for a set period of time within this range. They were then asked to recall the letters. Recall rates declined to near zero for delays of eighteen seconds, suggesting that information in STM is forgotten within this time frame. Figure 5.4 shows the results of the study across this range of delays.

Figure 5.4 STM Studies


Sources: From Peterson and Peterson (1959), Experiment 1. Photo from Jupiterimages/

Peterson and Peterson (1959) suggested that information decays from STM within eighteen seconds, as shown at the top of Figure 5.5. However, later studies have shown that another factor is more likely the cause of forgetting from STM: interference. When new information replaces old information in a memory store, this is known as retroactive interference. This occurs when new information effectively kicks old information out of STM (see the middle of Figure 5.5). Numerous studies have shown that retroactive interference occurs for information stored in STM. In fact, Waugh and Norman (1965) showed that the counting task in the Peterson and Peterson (1959) study likely interfered with the letter strings stored in STM, causing them to be forgotten. This is likely due to the way information is stored in STM. Encoding makes use of the different features of information (verbal, visual, meaning) to store that information in STM, but verbal features seem to be most important. Many studies have shown that subjects make more errors in STM retrieval based on similar verbal information than on other features of the information (e.g., confusing BAKE and RAKE from a list of words; e.g., Conrad, 1964; Hanson, 1990; Healy, 1974) and show higher recall for information that has a verbal feature than for information that does not have a verbal feature (e.g., Zhang & Simon, 1985). However, there is also evidence that visual and semantic (i.e., meaning-based, such as the connection between the items RAKE, LEAVES, and AUTUMN) features are also stored in STM (Brooks, 1968; Wickens, 1970). Feature coding in STM is discussed further in the section on working memory later in this chapter.

Retroactive interference: when new information interferes with the storage or retrieval of old information

Figure 5.5 Decay Versus Interference


Proactive interference has also been shown to cause forgetting from STM (see the bottom of Figure 5.5). This type of interference occurs when the old information already stored in STM keeps new information from being stored. Keppel and Underwood (1962) showed that in the Peterson and Peterson (1959) study, regardless of delay to recall, letter strings studied first had an advantage over letter strings studied later. This result suggests that proactive interference occurred such that early items in the list kept new information from being fully stored in STM, giving the early list items an advantage.

Today, researchers still debate the cause of forgetting from STM: decay or interference. Nairne and Neath (2013) suggest there is evidence for both processes to some extent, with decay responsible for a small amount of forgetting and interference responsible for most of the forgetting that occurs. They also suggest that interference comes in the form of temporal confusion: In order to recall items from a list just presented, you have to remember that it was on the most recent list and not on a list further in the past. Some studies (e.g., Neath & Knoedler, 1994; Turvey, Brick, & Osborn, 1970) have shown that changing the delay during the task can either decrease or increase recall, depending on whether the change in delay makes the items less or more temporally distinctive (Nairne & Neath, 2013). Thus, the cause of forgetting from STM is a topic still under investigation.

Proactive interference: when old information interferes with the storage or retrieval of new information

Long-Term Memory (LTM)

What did you have for breakfast yesterday? If you can recall this information, it is likely stored in your long-term memory (LTM). Unlike sensory and short-term memory, LTM appears to be an unlimited store of information. Studies (e.g., Bahrick, 1984) have shown that not only can we store information across our lifetimes in LTM, the amount of information that can be stored does not appear to have a limit. Thus, it is generally thought that LTM has both unlimited storage capacity and unlimited duration of storage. Also unlike STM, information is primarily stored according to its semantic features. This feature of LTM is shown in studies where meaning-based errors are easily obtained when related information is retrieved (Roediger & McDermott, 1995). What one can retrieve from LTM at a given time is limited. Retrieval of information from LTM depends on many factors that contribute to the context in which retrieval takes place. These factors are discussed further in Chapter 6 where we consider how to increase one’s retrieval from LTM. Common LTM tasks are also reviewed in Chapter 6.

Long-term memory: long-term (i.e., lifetime) storage of memory after some elaborative processing has occurred

Stop and Think

· 5.7. What is the capacity of STM? What can one do to increase this capacity?

· 5.8. Suppose you were trying to remember your nine-digit student ID number that you had just looked up on your web account in order to give it to someone over the phone. Your cell signal is not very good where your computer is located so you need to hold the number in your STM until you can make the call and report your number. How would you accomplish this task using your STM?

· 5.9. What is the most likely cause when information is lost from STM?

· 5.10. Describe some situations in your life in which you rely on your STM.

Types of LTM Memories

Three main types of memories can be stored in and retrieved from LTM: episodic memories (like what you had for breakfast yesterday), semantic memories (like what cognitive psychology means), and procedural memories (like how to make scrambled eggs). An episodic memory involves episodes from one’s daily experiences. Remembering what you did last Tuesday, the atmosphere of a party you went to last weekend, and the day you fell off the jungle gym in elementary school are all episodic memories. Some episodic memories are also autobiographical memories, because they allow us to do a kind of mental time traveling back to a particular episode in our lives. However, not all episodic memories are autobiographical. We can remember what we had for breakfast yesterday without feeling as if we have been mentally taken back to the point in time yesterday when we ate breakfast.

A semantic memory involves general knowledge we have but does not contain information about the time and place we learned that knowledge. You may know that Earth is the third closest planet to the sun, but you probably do not remember the day and place you learned that fact. Semantic memories contribute to many of our other cognitive abilities such as language (see Chapter 9) and concept formation (see Chapter 10). They also seem to be important in the formation of some types of false memories (see Chapter 7). The key difference between episodic and semantic memories is that episodic memories contain contextual information (e.g., time, place, mood) about the formation of the memory, whereas semantic memories do not contain this contextual information.

A procedural memory (sometimes called implicit memory—see Chapters 6 and 7 for more discussion of implicit memory tasks) involves “how to” instructions for skills and tasks. Knowing how to ride a bike or drive a car involves procedural memories once that skill is learned and can be performed somewhat automatically. These memories can be retrieved without us even intending to remember anything. Our abilities just seem to “flow” as we perform a task we know how to do, without much effort in retrieving the procedural steps. In fact, even amnesic individuals who lack the ability to intentionally retrieve episodic and semantic memories show retrieval of procedural memories (Warrington & Weiskrantz, 1970). For example, Clive Wearing, described in the introduction to this chapter, lost the ability to retrieve episodic and semantic memories (e.g., he could not remember where he was or why he was there), but he could still play the piano because his procedural memories could still be retrieved. We further discuss procedural memory later in this chapter and describe how it may be different from other types of memory at a neuropsychological level in Chapter 7.

Stop and Think

· 5.11. In what ways does LTM differ from STM?

· 5.12. Describe a memory of your own that fits each of the three types of LTM memory described in the previous section.

Episodic memory: memory for a specific episode or experience in one’s life

Semantic memory: memory for facts or knowledge

Procedural memory: memory for a skill or procedure

Brain function supports the distinctions between these types of memory (Moscovitch et al., 2007). As described earlier in this chapter, episodic memories (such as the scene of my dog, Daphne, lying in my office) are retrieved using the medial temporal lobe (MTL) areas, including the hippocampus, to pull back together the perceptual pieces of the memory from the cortical areas in which they are stored. However, semantic memory retrieval also relies on the MTL area, but the area activated by knowledge retrieval can depend on the type of knowledge being retrieved. Information seems to be stored in the area related to its use. For example, retrieval of motor information (e.g., a dog can run) will activate areas near the visual cortex areas that detect motion in the environment. The prefrontal cortex also seems to be more involved in retrieval of semantic than episodic memories. Procedural memories are thought to be retrieved using a different memory system altogether due to the abilities amnesics with MTL damage show in retrieving these types of memories. H. M., described in a famous case study in Chapter 2, was able to show improvement on procedural skills, even though he had no episodic memory for performing the tasks related to those skills in the past. Instead, procedural memories rely on the basal ganglia and its connections to the frontal lobe for retrieval.

The Working-Memory (WM) System

The description of the structure of memory as storage units based on the duration of storage that we have discussed so far in this chapter is one way of describing memory. However, there are other approaches to describing memory. For example, some (e.g., Squire, 2004) have described memory as a set of systems responsible for the encoding, storage, and retrieval of information. Working memory is one system that has been proposed for the control of memories that one encodes in, stores in, and retrieves from STM. You can think of STM as a fairly passive storage unit for information held over a short period of time. Working memory can be thought of as the system that controls the flow of information in and out of STM, keeping important information active in STM when it is needed and using the information to control the output from STM. In other words, the term working memory describes the system that controls the memories we are currently “working on” or “operating on” in our minds. As an example of the role of working memory in our lives, consider this scenario: You are biking down a busy walking path through your town. As you approach an intersection of the path and a busy street, you see another biker approaching from the opposite direction. There is also a person in front of you walking a dog that is on a leash but is rambunctious and veering across the path in an unpredictable manner. You also hear a nearby siren from the street you are approaching, but you do not see an emergency vehicle in the portion of the street you can see. To successfully navigate this scene, you need to be able to briefly store each piece of relevant information by focusing your attention on different parts of the scene and then processing the information such that you can anticipate where objects will be as you proceed on your bike. In this scenario, your working memory is controlling the input of visual and auditory information, coordinating that information to help you decide which way to steer your bike and where you should focus your attention at any given moment to achieve this task without crashing or being hit by cars passing in the street. If you think back to the way short-term memory was described earlier, with information coming in from sensory memory when it is the focus of attention and then either transferring on to long-term memory or being replaced by new information, this description does not seem complex enough to handle the bike-riding scenario. A more complex description of memory is needed to account for such behaviors.

Working memory: processing a unit of information that is the current focus of attention

Photo 5.4 The working-memory system controls our memories over the short term and our current focus of attention to allow us to perform complex tasks.

Image Mueller

Baddeley’s Model

Baddeley (Baddeley, 1992; Baddeley & Hitch, 1974) proposed the most prominent model of working memory. One thing that sets this model of working memory apart from the short-term memory storage we described earlier in the chapter is that it contains multiple storage subsystems for different types of information. It also proposes the existence of a central executive subsystem that controls the flow of information between the other storage subsystems and long-term memory and decides where one’s attention will be at any given moment. The primary storage subsystems in working memory are the visuospatial sketchpad and the phonological loop that hold visual and auditory information, respectively. In a newer version of the model, Baddeley (2000) added a fourth component that he called the episodic buffer, which acts as a temporary episodic storage subsystem and as a connection between working and long-term memory. Figure 5.6 illustrates his model of working memory.

Visuospatial Sketchpad

The visuospatial sketchpad is responsible for the storage of visual information in working memory. It acts as a type of dry-erase board for visual and spatial information that can be written on, stored for a brief time, erased, and rewritten on. However, as we will see in the description of studies that support the existence of the sketchpad, the information stored there can be moved around in the sketchpad and analyzed like a three-dimensional model. Much of the evidence for a separate subsystem for visuospatial information comes from studies where subjects are asked to perform two tasks at once. The researchers then look for interference in the tasks, depending on the type of tasks the subjects are asked to perform (e.g., two visuospatial tasks, versus one visuospatial and one verbal task). In other words, if lower task performance (i.e., more interference) is seen when both tasks involve the same type of information (two visual tasks) than when the two tasks involve different types of information (one visual task and one auditory task), then these results provide evidence that the working-memory system includes different subsystems for visual and auditory information.

Visuospatial sketchpad: the part of the working-memory system that holds visual and spatial codes of information

As an example of this type of study, we examine the methods used by Quinn and McConnell (1996) in their study. They asked subjects to remember a list of words either by verbally rehearsing the words (in their heads) or by forming a visual image of the words. While subjects were learning the words, they were also presented with a changing visual display (seemingly random visual block patterns) or no visual display. When the visual display was present, subjects who were told to visually imagine the words remembered fewer of the words than subjects who were told to verbally rehearse them. When no visual display was present, there was no effect on learning instruction. Figure 5.7 illustrates these results for the learning task and visual display conditions. These results showed that when irrelevant visual information is displayed during a visual learning task, subjects cannot perform the task as well as when they are doing a verbal learning task or when no irrelevant visual information is displayed. These results and others like them (e.g., Baddeley, 1998) show that when two tasks both rely on brief visual storage of information, they interfere with one another, supporting the notion of a separate storage subsystem in working memory for visuospatial information that has a limited capacity.

Figure 5.6 Baddeley’s (2000) Working-Memory Model


Other studies supporting the visuospatial sketchpad have shown that visuospatial figures can be manipulated mentally. For example, Shepard and Metzler (1971) asked subjects to judge whether two three-dimensional objects were the same or different (see Figure 5.8). The objects were rotated in space to different degrees. The researchers showed that the degree of rotation affected the time it took subjects to make the judgments (i.e., reaction time), such that each increment in degree of rotation increased the reaction time by the same amount. In other words, subjects were creating an image of the objects in the sketchpad subsystem of working memory and rotating those objects within the sketchpad to determine what they would look like when rotated to the same orientation as the comparison object. The more they had to rotate them mentally, the longer it took them to make their judgment. This is exactly the sort of task the visuospatial sketchpad is proposed to be useful for, and these results suggest that this subsystem of working memory is able to hold and manipulate this type of information.

Figure 5.7 Results of the Quinn and McConnell (1996) Study


Figure 5.8 Objects Used in the Shepard and Metzler (1971) Mental Rotation Study


Source: Shepard and Metzler (1971, figure 1).

Phonological Loop

The phonological loop is proposed to operate much like the visuospatial sketchpad but as a storage subsystem for verbal information. Verbal information is stored in a loop in this subsystem and then is replaced by new verbal information as it comes in. An articulatory control process in this subsystem allows rehearsal of the information to hold the information in the loop for a longer period of time. As described earlier for short-term memory, verbal codes (encoding information by its sounds) seem to be the dominant method of storing information for a short period of time; thus, the phonological loop has been the most heavily studied portion of the working-memory model. We have already described some evidence for the phonological loop in discussing short-term memory earlier in this chapter: More errors occur when recalling items that sound alike (e.g., C and T) than when recalling items that do not sound alike (e.g., C and X). This result occurs even when the items are presented visually because it is assumed that visual information involving language is automatically translated into verbal codes in working memory and stored in the phonological loop. Similar verbal codes (i.e., items that sound alike) can then become mixed up when recalling information stored in the phonological loop. This is known as the phonological similarity effect (Baddeley, 1998).

Phonological loop: the part of the working-memory system that holds auditory codes of information

In addition to the phonological similarity effect, studies have shown that having subjects repeat a word or phrase out loud while they learn from a written list reduces recall for those items. This is an effect known as articulatory suppression: articulatory rehearsal of list items is suppressed by the articulation of the irrelevant, repeated word. With both the repeated word and the items to be remembered stored in the phonological loop, it becomes overloaded and recall for the studied items is reduced. The list information cannot be rehearsed in the loop while it is also producing a verbal response. Studies by Peterson and Johnson (1971) and Baddeley, Lewis, and Vallar (1984) have shown these results for lists of letters and words, respectively.

The word length effect also supports the dominance of verbal coding in working memory and the existence of the phonological loop. The word length effect is seen when longer words (e.g., words with more syllables) show lower recall rates than shorter words.

Try this for yourself: Read over the following list of words. Then cover them up and try to recall them.

help, train, dream, gift, fight, blow, drive, brain, kite

How many could you remember? Probably about four to six of them, right? Now try a list with the same number of words.

helicopter, university, happily, hippopotamus, flowering, computer, fortify, opportunity, grocery

If you remembered fewer of the words in the second list, then you have illustrated the word length effect.

Baddeley, Thompson, and Buchanan (1975) showed this effect in their study comparing short-term recall for words with one syllable compared with words with five syllables. When the list contained five words, the lists with one-syllable words showed recall rates of almost 80 percent; however, the lists with five-syllable words showed recall rates of only about 30 percent. Figure 5.9 illustrates these results. Baddeley and his colleagues interpreted the results of their experiments as an indication that the time it takes to read a word out loud (i.e., the length of its verbal code) affects its recall. In other words, the word length effect is due to the longer words being forgotten more quickly because more time is passing when they are rehearsed in the phonological loop than for shorter words. Fewer of the longer words can be rehearsed before they are lost from short-term memory. This effect has been generalized to show that the length of time it takes to speak is related to recall span such that adults have a faster speech rate and higher recall span than children (Hulme, Thompson, Muir, & Lawrence, 1984). Further, recall span is higher for speakers of languages with faster speech rates (e.g., Chinese) than for speakers of languages with slower speech rates (e.g., Arabic or Welsh; Ellis & Hennelly, 1980; Naveh-Benjamin & Ayres, 1986).

Episodic Buffer

The episodic buffer is a subsystem of working memory proposed by Baddeley (2000) to handle the brief storage of episodic memories when the loop and/or sketchpad are otherwise engaged. For example, when performing articulatory suppression, one’s loop is completely engaged with the verbal repetition task and is unable to verbally store a list of items one wishes to remember. Yet recall of a list is not drastically impaired by articulatory suppression (Baddeley et al., 1984). Thus, the list items are being stored in another subsystem of working memory. Researchers have ruled out the sketchpad as a storage place for the list items during this task (Nairne & Neath, 2013); thus, a different storage subsystem is needed. Baddeley suggested that the episodic buffer serves in this role by briefly storing episodic memories with visual and verbal codes integrated from the other two storage subsystems. In other words, it can bind information with different codes (verbal, visual, semantic) to hold the combined information temporarily. It also serves as a link between working memory and long-term memory, allowing information stored in long-term memory to be used in the storage and retrieval of information in short-term memory.

Episodic buffer: the part of the working-memory system that holds episodic memories as an overflow for the phonological loop and visuospatial sketchpad

Figure 5.9 Results From Experiment 1 of Baddeley et al.’s (1975) Study for List Length of Five Items


Because it is the newest subsystem in the working-memory model, the episodic buffer and its functions have been tested by fewer studies than the other subsystems. The studies that have examined the episodic buffer have primarily focused on its binding function. Baddeley’s work (e.g., Baddeley, Hitch, & Allen, 2009) has shown that short-term memory for sentences is better than short-term memory for lists of words, indicating a role for binding of words using language knowledge and semantic information that increases the overall recall of words in sentences. Further, this effect did not depend on the amount of attention (based on verbal or visual interference) available for the tasks (see Figure 5.10 for their results). Thus, binding of features seems to occur automatically without requiring resources from the central executive and does not rely on the visuospatial sketchpad or phonological loop. Although Baddeley and his colleagues have begun testing the functions of the episodic buffer in recent studies (see Baddeley, 2012), it is clear that further work is needed to more fully describe the role of this subsystem in working memory.

Central Executive

If there is a manager of the working-memory system, it is the central executive. The central executive is the subsystem of working memory that controls the flow of information between the three storage subsystems described earlier, the flow of information between the episodic buffer and long-term memory, and which part of the system is the current focus of attention. In the biking example that opened this section, the central executive would be responsible for focusing your attention on the most important object and feature of that object at each moment as you move through the scene. This subsystem does not store information as do the other subsystems. Instead, it controls which information in the other subsystems is in our current focus of attention. However, as our attention is limited in what it can handle at any one time, the central executive also has a limited capacity in what it can control at any time. It is limited by the limits of our attention.

Central executive: the part of the working-memory system that controls the flow of information within the system and into long-term memory

Compared to the other subsystems of working memory, less research has been devoted specifically to examining the central executive subsystem of the Baddeley model due to its function as an attentional processing subsystem. However, numerous models of attention have been proposed (see Chapter 4) that could describe the functioning of the central executive component of working memory. For example, Baddeley (1998) has suggested that Norman and Shallice’s (1986) model of the control of action that includes a supervisory attentional system could describe the functioning of the central executive. In this model, many tasks are proposed to rely on automatic functioning (e.g., routines) with the supervisory attentional system coming into play when automatic functioning is not sufficient for a task. Baddeley argues that this model of attention can account for performance in tasks where the central executive would be expected to play a role (e.g., driving, playing chess, reading).

Stop and Think

· 5.13. Describe the four subsystems of Baddeley’s model of working memory. Which subsystem controls our focus of attention?

· 5.14. Which storage subsystem seems to be dominant in terms of features of information stored in working memory?

· 5.15. What role does the episodic buffer serve in working memory?

· 5.16. Describe two other perspectives on working memory besides the Baddeley model.

· 5.17. Describe some tasks from your life that involve your working memory. How might the working-memory model described earlier be involved in these tasks?

Figure 5.10 Results From Experiment 3 in Baddeley et al. (2009)


Beyond Baddeley’s Model

Although Baddeley’s is the most popular model for working memory and has been tested more than other models, some researchers have suggested other ways to conceptualize working memory. For example, Cowan (1999) has suggested that instead of being a separate system of memory as Baddeley’s model proposes, working memory is simply the part of long-term memory that is currently activated in our attention. In other words, long-term memory is the main memory system with working memory operating on a portion of long-term memory currently active in our attention. Another approach to describing working memory is through neurobiology. Jonides and colleagues (2008) examined the neural activity that accompanies the encoding, storage, and retrieval of information over the short term, with an emphasis on brain activity that occurs when information is the focus of attention and binding the features of the information when it is stored. The researchers rely on studies using the techniques of cognitive neuroscience (see Chapter 2) to support their approach to working memory. Thus, the study of working memory is being conducted from multiple perspectives.

Memory Overview

In this chapter, we discussed the processes of encoding, storage, and retrieval from memory and the approaches researchers have taken in their study of these processes. In this discussion, we identified several forms of memory, and some additional forms of memory (e.g., implicit, autobiographical, prospective) will be presented in the next chapter. However, there is no clear agreement among researchers on how many types of memory there are. Some of the forms of memory described in this chapter are similar enough to one another that they may not represent distinct forms of memory (e.g., short-term and working memory). One way to distinguish different forms of memory is to determine the brain systems responsible for them. We’ll discuss this approach further in Chapter 7 in the section on amnesia. It is also possible that researchers have yet to identify additional forms of memory that are distinct from the forms presented in this text.

Models of retrieval have also been developed as a way to describe the process of retrieval that occurs in different memory tasks. Some models propose a single retrieval mechanism from memory for all forms of memory, whereas other models focus on a specific type of memory and the processes involved in retrieval for a certain type of task. Thus, the process (or processes) of retrieval is an ongoing topic of study for memory researchers. In Chapter 6, we focus on the types of tasks that measure different types of memory retrieval and how different aspects of encoding and retrieval can be used to increase memory retrieval.

Thinking About Research

As you read the following summary of a research study in psychology, think about the following questions:

1. Which aspect of Baddeley’s (2000) working memory model does this study seem to address? What do the results tell us about this part of working memory?

2. What type of research design are the researchers using in this study? Explain your answer. (Hint: Review the Research Methodologies section in Chapter 1 for help in answering this question and Question 3.)

3. What are some possible controls the researchers likely included in this study? Why are these controls important?

4. What are some practical implications of the results of this study?

Study Reference

Xu, Y., & Franconeri, S. L. (2015). Capacity for visual features in mental rotation. Psychological Science, 26, 1241—1251.

Note: Experiment 1a of this study is presented.

Purpose of the study: In this study, the researchers examined mental rotation abilities to address the question of how visual information can be held for an object while mentally rotating it. More specifically, they measured identification accuracy for parts of a rotating cross that the participants were asked to mentally rotate compared with accuracy for parts of a non-rotated cross (see Figure 5.11).

Method of the study: Twelve paid participants completed the study. On each trial, participants first saw either a color wheel that did not move or a rotating color wheel along with a mechanical sound to simulate movement of the wheel. On rotation trials (the moving wheel) they were asked to remember the direction and rate of the rotation. They then saw the colored cross in a particular orientation for 500 milliseconds. On non-rotating trials, this was followed by a delay with a blank screen for 800, 1,600, or 2,400 milliseconds. On rotating trials, they were asked to imagine the cross rotating at the same rate and in the same direction as the colored wheel they saw earlier while viewing a blank screen for 800, 1,600, or 2,400 milliseconds. Finally, they viewed the colored cross in a particular orientation and had to judge if the cross shown either matched the original cross (non-rotating trials) or matched the cross that was mentally rotated (rotating trials). Figure 5.11 shows the stimuli and procedure sequence for this study for non-rotating and rotating trials. While performing this task, participants also completed a concurrent rehearsal task of four consonants they were to remember to block any verbal contribution to the mental rotation task.

Figure 5.11 Stimuli and Trial Sequence for the Xu and Franconeri (2015) Study


SOURCE: Adapted from Xu, Y., & Franconeri, S. L. (2015). Capacity for visual features in mental rotation. Psychological Science, 26(8), 1241—1251.

Results of the study: The researchers calculated a measure of the number of features that were correctly identified in the rotated and non-rotated objects. Capacity was found to be double in size for the non-rotated compared with the rotated objects. These results are displayed in Figure 5.12.

Conclusions of the study: The researchers concluded that mental rotation significantly reduces the number of features that can be remembered for objects in working memory.

Figure 5.12 Mean Feature Capacity by Condition for the Xu and Franconeri (2015) Study


SOURCE: Adapted from Xu, Y., & Franconeri, S. L. (2015). Capacity for visual features in mental rotation. Psychological Science, 26(8), 1241—1251.

Chapter Review


· Is memory a process, a structure, or a system?

Memory has been thought of as both a process and a structure. Researchers have viewed memory in terms of processes (encoding, storage, and retrieval), structural storage units (sensory, short-term, and long-term memory), and systems (working-memory system with multiple subsystems).

· How many different types of memory are there?

There is no clear answer to this question, as it is unclear which types of memory are distinct from other types. However, researchers have attempted to identify several different types of memory: memory based on duration (short-term vs. long-term memory) and memory based on content (episodic, semantic, and procedural memory); both were discussed in this chapter. In the next chapter, we will discuss memory based on retrieval task (recall and recognition), memory based on reference to the self (autobiographical memory), memory based on vivid details and emotional context (flashbulb memory), memory based on intentionality of retrieval (explicit vs. implicit memory), and memory for future tasks (prospective memory). However, this list of memory types is based on current ideas of how memory can be classified and not meant to be inclusive of all forms of memory that researchers have studied in the past or will study in the future. The answer to this question remains under investigation.

· Are there differences in the ways we store and retrieve memories based on how old the memories are?

Yes. There are important differences in memories we store for the short term and memories stored over the long term. The main distinction between these types of memories is the duration of storage: less than a minute for short-term memories and a lifetime for long-term memories. In addition, short-term memories seem to be coded primarily with verbal codes, and long-term memories seem to be coded primarily with semantic codes. Finally, the capacity of short-term memory seems to be limited (about five to nine chunks of information), whereas long-term memory seems to have an unlimited capacity.

· What kind of memory helps us focus on a task?

Working memory involves information about a task currently in our focus of attention. Thus, it aids in the completion of tasks we are currently attending to, while also helping us keep track of other things in our environment and ignore things that are irrelevant.

· What are the limits of our memory?

In some cases, the limits of memory are based on our limits of attention in terms of what we can encode effectively and focus on for appropriate cues for retrieval. Over the short term, our attention limits influence what we can focus on in working memory (or store in STM). Over the long term, we seem to be able to store unlimited amounts of information, but we are limited in what we can retrieve at any given time.

Chapter Quiz

1. Enter the letter for the memory term next to the example below that illustrates that form of memory.

1. semantic memory

2. episodic memory

3. procedural memory

§ ___ after years without practice you pick up a golf club and make an excellent drive

§ ___ you know that the capital city of China is Beijing

§ ___ you remember the time you went with your friends to the movies to see The Hunger Games

2. Which memory storage unit in the modal model of memory holds information for a second or two as raw sensory information?

1. working memory

2. long-term memory

3. short-term memory

4. sensory memory

3. Which subsystem of the working-memory system controls the focus of attention?

1. the episodic buffer

2. the central executive

3. the phonological loop

4. the visuospatial sketchpad

4. Which subsystem of the working-memory system allows for rehearsal of information held for the short term?

1. the episodic buffer

2. the central executive

3. the phonological loop

4. the visuospatial sketchpad

5. Which of the following is an example of a procedural memory?

1. remembering what you had for breakfast yesterday

2. remembering how to make breakfast

3. remembering to have breakfast before you leave the house

4. remembering what the word breakfast means

6. Describe the types of memory errors one is likely to make if one studies and recalls the following list—happy, game, honey, trust, lame, bee.

1. when recall occurs after thirty seconds

2. when recall occurs after twenty-four hours

7. Describe the role of the central executive in Baddeley’s model of working memory.

8. Provide examples of both proactive and retroactive interference.

9. Explain how you could remember the following sequence of numbers for a short time using chunking: 1 8 1 2 2 0 0 0 2 0 1 8 1 7 7 6

Key Terms

· Central executive 124

· Chunking 113

· Encoding 107

· Episodic buffer 123

· Episodic memory 117

· Long-term memory 116

· Partial-report method 110

· Phonological loop 122

· Proactive interference 116

· Procedural memory 117

· Retrieval 107

· Retroactive interference 115

· Semantic memory 117

· Sensory memory 110

· Short-term memory 112

· Storage 107

· Visuospatial sketchpad 119

· Working memory 118

Stop and Think Answers

· 5.1. Describe the three primary processes of memory.

Encoding is the process of getting information into memory. Storage is the process by which information is held in memory. Retrieval is the process by which information is remembered.

· 5.2. List the three hypothetical storage structures of memory from the shortest to the longest storage.

Shortest: sensory memory; intermediate: short-term memory; longest: long-term memory

· 5.3. Consider different ways in which you encode information you learn in class (e.g., visually, aurally). How effective do you think each of these encoding processes is for storing information in long-term memory?

Answers will vary.

· 5.4. Explain how the partial-report method allows researchers to more accurately estimate the capacity of sensory memory than a whole-report method.

The partial-report method allows for a report of a smaller amount of information than the whole set of stimuli presented. Because information is lost from sensory memory so quickly, it is difficult for one to report what is stored in sensory memory before it is lost. The partial-report method allows researchers to estimate how much information is stored before it is lost by extrapolating from the part the subject is asked to report to the whole set of stimuli presented.

· 5.5. According to the research in this area, what is the duration of sensory memories?

The duration of sensory memory is believed to be about one second for visual information and a little longer (about four seconds) for auditory information. However, due to the differences in how these modes of stimuli are presented, it is difficult to know if there is a different duration for visual and auditory information or if the differences found in research are caused by modality of the information.

· 5.6. Research in sensory memory for senses other than vision and audition is scarce. Imagine that you are researching olfactory (sense of smell) sensory memory to contribute to the gap in the research in this area. Describe a study you might design using the partial-report method to study olfactory sensory memory. What are some of the limitations of this method for this type of sensory memory?

Answers will vary, but the limiting factor is allowing subjects to report the information stored in sensory memory before it is lost. It is difficult to apply the partial report to other senses, which is one reason there has been less research done on the other senses.

· 5.7. What is the capacity of STM? What can one do to increase this capacity?

The capacity of STM seems to be about five to nine chunks of information. The capacity can be increased with chunking (i.e., organizing the information into fewer units according to meaning). For example, more letters can be stored in STM if they are chunked into words when they are encoded.

· 5.8. Suppose you were trying to remember your nine-digit student ID number that you had just looked up on your web account in order to give to someone over the phone. Your cell signal is not very good where your computer is located so you need to hold the number in your STM until you can make the call and report your number. How would you accomplish this task using your STM?

Answers will vary. Based on results from studies using the Brown-Peterson method, information can be stored in STM for about twenty seconds. You can increase the duration of storage of information in STM by rehearsing the information to keep it in your focus of attention.

· 5.9. What is the most likely cause when information is lost from STM?

Interference is the most likely cause (either proactive or retroactive interference).

· 5.10. Describe some situations in your life in which you rely on your STM.

Answers will vary.

· 5.11. In what ways does LTM differ from STM?

Storage duration and capacity in LTM appears to be unlimited, whereas it is clearly limited in STM. In addition, the primary mode of storage of information in LTM seems to be the meaning of the information, whereas verbal coding is the dominant storage mode in STM.

· 5.12. Describe a memory of your own that fits each of the three types of LTM memory described in the previous section.

Answers will vary. Episodic memories are for episodes, semantic memories are for facts and knowledge, and procedural memories are for skills memories.

· 5.13. Describe the four subsystems of Baddeley’s model of working memory. Which subsystem controls our focus of attention?

The phonological loop (verbal information) and visuospatial sketchpad (visual information) serve as storage units of information in working memory. The episodic buffer stores episodic information and connects with LTM. The central executive acts as the control system to determine what our attention is currently focused on.

· 5.14. Which storage subsystem seems to be dominant in terms of features of information stored in working memory?

The phonological loop appears to be the dominant subsystem for storing information.

· 5.15. What role does the episodic buffer serve in working memory?

The episodic buffer stores episodic information and connects with LTM.

· 5.16. Describe two other perspectives on working memory besides the Baddeley model.

Other perspectives include describing working memory as the activated portion of LTM and describing working memory through the brain activity that accompanies encoding, storage, and retrieval of memories in the short term.

· 5.17. Describe some tasks from your life that involve your working memory. How might the working-memory model described earlier be involved in these tasks?

Anything that involves one’s current focus of attention will qualify for working memory. Answers will vary.

Student Study Site


SAGE edge offers a robust online environment featuring an impressive array of free tools and resources for review, study, and further exploration, keeping both instructors and students on the cutting edge of teaching and learning.



JIMIN LAI/AFP/Getty Images