Memory

Neuropsychological, Imaging and Psychopharmacological Perspectives

Gérard Emilien, Cécile Durlach, Elena Antoniadis, Martial Van der Linden, Jean-Marie Maloteaux

• Publisher: Routledge
• Year: 2004
• Print ISBN: 9781841693705
• eBook ISBN: 9781135424428
• Edition: 1
• Topic: Psychology
• Subtopic: Cognitive Neuroscience & Neuropsychology
• Index: Psychology

Memory: Neuropsychological, Imaging and Psychopharmacological Perspectives reviews critically the impact of recent neuropsychological and biological discoveries on our understanding of human memory and its pathology. Too often, insights from clinical, neurological and psychopharmacological fields have remained isolated and mutually unintelligible. Therefore the first part of this book provides both clinicians and neuroscientists with a broad view of the neuropsychology of memory, and the psychobiological processes it involves, including recent advances from imaging technology and psychopharmacology research. In the second part the authors go on to cover a comprehensive range of memory assessments, dysfunctions, impairments and treatments. This compendium of current research findings will prove an invaluable resource for anyone studying, researching or practising in the field of memory and its disorders.

Part I

Neuropsychology and Pharmacology of Memory

1

What is Memory?

INTRODUCTION

In memory research, the initial impact of an experience on the brain has been called an engram (Schacter, 1996). Trace theory describes an encoding process in which the engram or memory trace has both a gist (a general notion) and specific details (Schacter, 1996). With time, the details of an experience may begin to fade away and become less tightly bound together. However, the gist may remain quite accurate and easily accessible for retrieval. When trying to retrieve an original memory, one may be calling up the gist at first and then later trying to reconstruct the details. This reconstruction process may be profoundly influenced by the present environment, the questioning context itself, and other factors such as current emotions (Christianson, 1991).

Remembering past events involves a large number of processes and their identification and interaction continues as a primary goal of today’s memory research. One major set of these processes concerns “ecphory” (recovery of stored information); another has to do with what is called episodic “retrieval mode” or REMO (Nyberg et al., 1995; Tulving, 1983). REMO refers to a neurocognitive set or state, and is a basic and pivotal necessary condition for remembering past events. Memory is an active process in which even the most concrete experiences are actually dynamic representational processes. Remembering is not merely the reactivation of an old engram. It is the reconstruction of a new neural-net profile with features of the old engram and elements of memory from other experiences, as well as influences from the present state of mind. This is particularly important for clinicians when interviewing patients about past events, including trauma. The gist of a situation may be recalled with great accuracy but the details may be more subject to bias from ongoing experiences.

Considering the importance of context in memory, we may define context as information associated with a specific memory that allows differentiation of that memory from other memories. Intrinsic context refers to features that are an integral part of the stimulus itself. In a face, for example, intrinsic contextual features would be eye colour, hair length, etc. In the case of words, intrinsic context would relate to the particular meaning extracted from the word at the time of learning. Extrinsic context corresponds to those features that include time of encounter and surroundings. Studies of memory dysfunctions and amnesia have concentrated on memory for extrinsic context.

Two fundamental ways of gaining access to past experiences may be distinguished: recollection of contextual details surrounding a previous encounter with a stimulus (source memory) and a general sense of familiarity that is sufficient to determine whether the stimulus was previously encountered even though contextual recollection is absent (item memory) (Atkinson & Juola, 1974; Tulving, 1985; Yonelinas, 1994). Item memory combined with source memory failure is a common experience in everyday life and plays an important role in such problems as faulty eyewitness identification, when a face perceived as familiar is assigned the wrong context. This distinction has implications for understanding the neural organisation of memory. Extensive neuropsychological research suggests that source memory can be dissociated from memory for particular items (Schacter et al., 1984; Shimamura et al., 1991).

Memory is not situated only in one structure in the brain; it is a biological and psychological phenomenon involving the association of several brain systems working together. Although it was initially believed that memory was a general property of the cerebral cortex as a whole, evidence from patients with temporal lobe damage was instrumental in establishing that there are different types of memory localised in the brain. Residual memory has been observed in these patients for both associative and non-associative forms of implicit memory. Despite the deficits in explicit memory tasks, these patients retain various forms of reflexive learning, including habituation, sensitisation, and classical conditioning, all forms that have an automatic quality. Recent studies have highlighted the differences in mechanisms and determinants between declarative and procedural memory (Tulving & Schacter, 1990) (see Figure 1.1 & 1.1).

In human memory, the mechanisms of forgetting are not so clear. Early in the cognitive revolution, one view was that STM is vulnerable to decay (Brown, 1958; Peterson & Peterson, 1959). Information “evaporates” with time unless it is actively maintained. Another view was that forgetting is caused by interference from distracting elements (Keppel & Underwood, 1962). Interference has dominated the memory literature but decay has not entirely been forgotten. Support for decay comes from evidence of forgetting in the absence of interference (Baddeley & Scott, 1971). Interference certainly remains a potent source of forgetting even at long delays. Decay and interference are probably functionally related (Altmann & Gray, 2002). If a target decays, it will interfere less with future targets.

The ways that we encode, store, retrieve, and forget information provide clues to what declarative memory is and how it functions. Declarative memory is memory for events, facts, words, faces, music, all the various pieces of knowledge that we have acquired during a lifetime of experience and learning. It is knowledge that can potentially be declared and brought to mind as a verbal proposition or as a mental image. Declarative memory encodes information about autobiographical events as well as factual knowledge. Its formation depends on cognitive processes such as evaluation, comparison, and inference. Subjects are aware of the source and context of remembered information. Declarative memory can be recalled by a deliberate act of recollecting. It often can be concisely expressed in declarative statements such as “Last winter I visited my aunt in the countryside” (autobiographical event) or “All that glitters is not gold” (factual knowledge). It critically depends on structures in the medial temporal lobe. It is also called explicit or conscious memory. Research also suggests that episodic or explicit memory is characterised by the experience of specific features (e.g., colour, shape, semantic, time, location, and emotion) that comprise the event (Johnson & Raye, 1981). Thus, episodic memory depends on encoding multiple features and associations among them. Furthermore, reflective activities (e.g., rehearsal, organisation, semantic judgements, and imagery) contribute to establishing such complex event representations. In contrast, procedural memory, which includes the ability to learn motor skills, does not. Procedural memory processes are, therefore, perhaps best characterised as “non-conscious memory” in order to distinguish them from repressed or dynamically unconscious memory processes (Clyman, 1991). They are specialised memory systems for processing patterns of perception, emotion, and action without the mediation of conscious attention or recollection. The distinctions between declarative and procedural memory are most apparent in amnesic patients. Amnesic individuals with bilateral medial temporal lobe lesions can learn procedural skills, whereas word list learning cannot be recalled. Both declarative and procedural memory, however, are dependent on frontal cortex activity. Tests of declarative memory often require more than simple determination of whether a stimulus is familiar. Older people tend to be impaired in explicit but not in implicit memory tests (Bäckman et al., 1997). Older people with dementia lose the ability to encode or recall verbal declarative memories but may continue to be able to learn motor tasks such as solving a difficult puzzle. These patients can demonstrate improvement in performing a puzzle task from session to session, yet they cannot recall that they even attended the puzzle sessions or recognise the examiner.

TABLE 1.1

Organisation of memory

FIGURE 1.1

Schema for organising various types of memories.

Memory tasks such as free recall require planning to perform the task successfully. This sort of memory is termed strategic memory because it involves the use of strategies beyond simply deciding on the familiarity of a presented cue. On the other hand, source memory refers to the origin of information, that is, when, where, and from whom one may have acquired a particular piece of knowledge (Lindsay et al., 1991a). This type of memory has been hypothesised to be dependent on frontal lobe function. Much of the psychological research on memory processes in recent years has been based on a widely used information-processing model, referred to as the “multimodal” memory store (Atkinson & Shiffrin, 1971). Incoming sensory information is initially held in preconscious, transient stores. Any information that is not lost is rapidly transferred (within milliseconds) to “short-term” or “primary” memory, where it is available for conscious processing, typically for periods of 10 to 20 seconds. Memory can be categorised as short-term (STM) and long-term (LTM). STM has limited capacity (only a few items), persists for a short time, and is believed to be related to a variety of short-term changes such as augmentation, synaptic facilitation, and post-tetanic potentiation (Zucker, 1989). STM and working memory are both central constructs in modern theories of memory and learning. STM refers to information in LTM that is activated above some threshold (Cowan, 1988, 1995). Activated information rapidly returns to an inactive state unless it becomes the focus of a limited-capacity attentional process. Working memory includes STM as well as the attentional processes used to keep some STM contents in an activated state. A study aiming to determine whether STM and working memory could be distinguished was performed in children aged 7–13 years (Kail & Hall, 2001). It was observed that both exploratory and confirmatory factor analyses distinguished STM tasks from working memory tasks. Performance on working memory tasks was related to word-decoding skill but performance on STM tasks was not. The existing evidence, though scanty, seems consistent with the hypothesis that STM and working memory are distinct but related. Working memory seems to play a greater role than STM in higher-order cognitive processes. Performance on both STM and working memory tasks were associated with age-related increases in processing speed.

STM can be transformed into a more permanent storage of information (LTM) in the cortex. Such transformation into LTM is achieved through a more elaborate encoding process known as “depth of processing”. This refers to the amount of effort and cognitive capacity employed to treat information, and the number and strength of associations that are thereby forged between the data to be learned and knowledge already in memory. This can be established with an elaborate encoding process that makes new information meaningful and therefore easy to consolidate. The greater the depth of processing, the greater the ability to recall learned information. LTM can be divided into an intermediate form, which is more easily disrupted, and a long-term form, which is resistant to disruption. In general, the formation of LTM takes several hours, during which time memories rely on short-term systems (Izquierdo & Medina, 1997). LTM is presumably mediated by persistent functional and structural changes in the brain. It is seemingly unlimited in its capacity and can retain many thousands of concepts and patterns, sometimes for a lifetime.

Another, slightly different model of LTM suggests two different facets of memory—episodic and semantic (Schacter & Tulving, 1994; Tulving, 1987). Episodic memory refers to information that is remembered within a particular temporal or spatial context (Tulving, 1983). It is autobiographi- cal, that is, the recollection of information linked to an individual’s personal experiences. It is called “episodic” because it is composed of a series of events (or an episode), which are sequentially ordered in time. Its time-linked nature is an important aspect and may comprise both reflections into the past or looking forward to the future (Tulving, 1989). Thus, episodic memory allows the mind to have a sense of space and time and to remember where things are and when they were there.

Semantic memory, on the contrary, is impersonal and not time-linked. It comprises an individual’s repository of general information. The archetypal form of semantic memory is the process of using language, which calls on the individual’s mental vocabulary and repertoire of knowledge of syntactical and semantic rules governing grammar, word order, meaning, and usage. On the other hand, episodic memory is more “plastic” than semantic memory. It may be affected by the process of recollection itself (e.g., by retrieval) or by a variety of factors that produce interference (e.g., distractors, anxiety) (Tulving & Schacter, 1990). Episodic and semantic memory share important features. They may for certain purposes be grouped together under the category of declarative memory (Tulving, 1983). Retrieval from procedural, perceptual representation system (PRS), and semantic memory is classified as implicit; that from STM and episodic memory is classified as explicit.

One useful concept of memory organisation is what some cognitive psychologists call a “schema”. A schema is any pattern of relationships among data stored in memory. It is any set of nodes and links between them in the spider-web of memory that hang together so strongly that they can be retrieved and used more or less as a single unit. Theories of memory organisation suggest that activity knowledge organises autobiographical memory globally. According to these views, memories that share a participant, location, or time are only organised together if they also share an activity. If they do not, they are nested within their respective activity organisations locally rather than being organised together globally. However, this view is contradicted by a laboratory study that investigated the extent to which non-activity event characteristics organise event memories globally and provide direct access to these memories unmediated by activity (Lancaster & Barsalou, 1997). It was reported that people organise event memories globally in non-activity clusters, cross-classify events into multiple organisations, and pivot between activity and non-activity clusters. These results demonstrate the ability of individuals to organise events and also suggest that they may cross-classify event memories simultaneously into multiple global organisations. However, the representational and processing mechanisms that make such flexible performance possible need to be further understood.

Memories can also be divided into two additional categories: associative and non-associative. While both these categories cover many different subtypes, the essential difference is that in associative learning an animal learns the predictive nature of one environmental stimulus for another. The classic example of associative learning is Pavlovian associative conditioning, where Pavlov’s dogs learned that a sound reliably predicted the delivery of food. The memorability of an event increases when that event is related to preexisting knowledge, to existing beliefs or expectations, and/or other events at the time of encoding. Memory retrieval is enhanced by cues that are similar to the information supplied at the time of encoding. In addition, the memorability of an event increases when the event is relevant to expectations and beliefs about the event (Kihlstrom, 1994). Stressful events are often more memorable than ordinary daily life events. Cognitive theories have postulated that association and planning depends on the activation of representation units called “scripts” (Schank, 1982). Scripts are thought of as knowledge structures used for the representation of experienced events. They contain information about the temporal ordering, duration, and relative importance of events. Their structure is supposed to be similar to lexical or semantic knowledge representations in which individual items are linked by associative rules to form a network (Grafman, 1994).

Distinctions are also often made between verbal and spatial memory (i.e., memory for words versus spatial locations or pictures, respectively). More recently this distinction has been modified to reflect modality of input or an auditory versus visual distinction (The Psychological Corporation, 1997). As for the clinical relevance of these distinctions, there is a strong tendency to concentrate assessment on episodic and declarative memory rather than semantic and procedural memory, even though assessing these latter memory types can be very diagnostic in some memory disorders (Butters et al., 1988).

There are three ways in which information may be learned or committed to memory: by rote, assimilation, or use of mnemonic device. In the by rote approach, material to be learned is frequently repeated verbally so that it can later be repeated from memory without the use of any memory aids. When information is learned by rote, it forms a separate schema not closely interwoven with previously held information. That is, the mental processing adds little by way of elaboration to the new information, and that adds little to the elaboration of existing schemata. Learning by rote seems to be the least efficient way of remembering.

Through the assimilation method, the information is learned when its structure or substance fits into some memory schema already possessed by the learner. The new information is assimilated to the existing schema and can be retrieved readily by first accessing the existing schema, which may reconstruct it. Assimilation involves learning by comprehension. It is therefore a desirable method but it can only be used to learn information that is somehow related to one’s previous experience.

Lastly, information may also be memorised by use of mnemonic techniques. A mnemonic device is any means of organisation or encoding information for the purpose of making it easier to remember. Mnemonic devices are useful for remembering information that does not fit any appropriate conceptual structure or schema already in memory. They work by providing a simple, artificial structure to which the information to be learned is then linked. The mnemonic device supplies the mental “file categories” that ensure retrievability of information. To remember, first recall the mnemonic device, then access the desired information.

SOME CONCEPTS OF MEMORY

Pairs of features that can be used to classify memory tasks and interventions are effortful versus automatic processing of information and the engagement of explicit versus implicit memory systems. The distinction between effortful versus effortless or automatic learning has been vigorously researched (Hasher & Zacks, 1979; Kausler, 1990; Parkin & Russo, 1990). Theoretical approaches to the study of memory make a distinction between automatic and controlled processes in remembering (Jacoby, 1991, 1998, 1999). Controlled processes are characterised by the intentional and conscious recollection of previously learned material. In contrast, automatic influences operate without intention, and lead to demonstrations of learning in the absence of any awareness of the learning episode. Initially, controlled and automatic memory processes were identified with remembering in direct and indirect memory tests (Ste-Marie et al., 1996). An example of a direct test is the traditional free recall procedure where participants ar...

Table of contents

• COVER PAGE
• TITLE PAGE
• COPYRIGHT PAGE
• INTRODUCTION
• PART I: NEUROPSYCHOLOGY AND PHARMACOLOGY OF MEMORY
• PART II: PATHOLOGY OF MEMORY
• REFERENCES