《认知神经科学》课程教学资源（参考文献）[Milner, B., Squire, L. R., & Kandel, E. R.（1998）]Cognitive neuroscience and the study of memory.pdf

Neuron, Vol. 20, 445–468, March, 1998, Copyright 1998 by Cell Press Cognitive Neuroscience Review and the Study of Memory that eventually led to the independent discipline of experimental psychology. In its early years, experimental psychology was concerned primarily with the study of Brenda Milner,* Larry R. Squire,† and Eric R. Kandel‡§ *Montreal Neurologic Institute Montreal, Quebec H3A 2B4 sensation, but by the turn of the century the interests Canada of psychologists turned to behavior itself—learning, †Veterans Affairs Medical Center memory, attention, perception, and voluntary action. San Diego, California 92161 The development of simple experimental methods for and University of California studying learning and memory—first in humans by HerSan Diego, California 92093 mann Ebbinghaus in1885 and a few years later inexperi- ‡Center for Neurobiology and Behavior mental animals by Ivan Pavlov and Edgar Thorndike— College of Physicians and Surgeons led to a rigorous empirical school of psychology called Columbia University behaviorism. Behaviorists, notably James B. Watson Howard Hughes Medical Institute and Burrhus F. Skinner, argued that behavior could be New York, New York 10032 studied with the precision achieved in the physical sciences, but only if students of behavior abandoned speculation about what goes on in the mind (the brain) and The neurosciences have grown rapidly over the last half focused instead on observable aspects of behavior. For behaviorists, unobservable mental processes, espe- century. This growth has been stimulated by two impor- cially abstractions like perception, selective attention, tant developments. First, molecular biology has trans- and memory, were deemed inaccessible to scientific formed cellular neurobiology and has led to a new con- study. Instead, behaviorists concentrated on examin- ceptual framework for signaling, a molecular framework ing—objectively and precisely—the relationship be- that encompasses not only signaling in nerve cells but tween specific physical stimuli and observable re- in all the cells of the body. Second, work on brain and sponses in intact animals. Their early successes in cognition, which was traditionally associated with a number of different disciplines, has merged into a single rigorously studying simple forms of behavior, including discipline: cognitive neuroscience. This has provided a learning, encouraged them to treat all processes that intervene between the stimulus (input) and behavior new framework for the study of memory, perception, (output) as irrelevant to a scientific study of behavior. action,language, and perhaps even conscious awareness. Thus, behaviorism largely ignored mental processes. As In this review, we will consider the second development by focusing on one aspect of cognitive neuro- a result, the science of behavior was defined in terms of the limited techniques used to study it. This emphasis science: recent progress in memory research. In so do- reduced the domain of experimental psychology to a ing, we also want to consider the broader question: to what degree can these two independent and disparate restricted set of problems, and it excluded from study some of the most fascinating features of mental life. strands—molecular neurobiology and cognitive neuroscience—be united? Can molecular biology enlighten By the 1960s, it was not difficult for the founders of cognitive psychology—George Miller, Ulric Neisser, the study of cognitive processes, such as learning and memory, as it has other areas of biology, such as devel- Herbert Simon, and others—to convince the scientific opment? In turn, can cognitive neuroscience define community of the narrowness of behaviorism. These novel phenomena that will lead to a completely new set early cognitive psychologists, building on the earlier eviof molecular mechanisms and insights? dence from Gestalt psychology, European neurology, and work by the British psychologist Frederic Bartlett, The Emergence of Cognitive Neuroscience sought to demonstrate that our knowledge of the world Cognitive neuroscience originated in two disciplines: in is based on our biological apparatus for perceiving the psychology, in the development of rigorous methods world, and that perception is a constructive process for analyzing behavior and cognition, and in systems dependent not only on the information inherent in a neurobiology, in the effort to understand the structure stimulus but also on the mental processing of the perand function of neuronal circuits of the sensory and ceiver. Thus, cognitive psychology was concerned not motor systems of the brain. The fusion of these two simply with specifying the input and output for a particudisciplines was facilitated as well by the emergence of lar behaviorbut also with analyzing the processby which a coherent neuroscience—an interdisciplinary approach sensory information is transformed into perception and to the nervous system that encouraged the idea that the action—that is, with evaluating how a stimulus leads to techniques and concepts of neurobiology and systems a particular behavioral response. In redirecting scientific neuroscience might be usefully applied to the analysis attention to mental operations, cognitive psychologists of cognition. focused on information processing, on the flow of senUntil the beginning of the nineteenth century, the study sory information from sensory receptors to its eventual of normal mental activity was a part of philosophy, and use in memory and action. It was implicit in the cognitive the chief method for understanding the mind was intro- approach to behavior that each perceptual or motor act spection. By the middle of the nineteenth century, intro- has an internal representation in the brain: a representaspection began to give way to experimental approaches tion of information in patterns of neural activity. Once cognitive psychologists acknowledged that internal representations are an essential component of §To whom correspondence should be addressed

Neuron 446 behavior, they had to come to grips with the fact that of information about the organization and anatomy of most mental processes were still largely inaccessible higher functions, including memory. Lesionstudies have to experimental analysis. Without direct access to the shown that cognition is not unitary but that there are neural substrates of internal representations it was diffi- several cognitive systems, each with independent inforcult, if not impossible, to understand the path from per- mation-processing modules. For example, the visual ception to action. At about this time, the work of Vernon system of primates, a prototypicalcognitive system, has Mountcastle on somatic sensation, David Hubel and specialized anatomical pathways for processing inforTorsten Wiesel on vision, and Edward Evarts on the mation about color, form, and movement. control of movement inaugurated the neuronal analysis Finally, computational science has made a distinctive of perception and voluntary action. Moreover, during contribution to cognitive neuroscience. Computers made the 1970s, Evarts and Mountcastle developed tech- it possible to model the activity of large populations of niques for studying the activity of single cells in the neurons and to begin to test ideas about how specific brains of awake, behaving monkeys. In their hands, and components of the brain contribute to particular cogniin work that followed by Robert Wurtz, Apostolos Geor- tive processes. To understand the neural organization gopoulos, William Newsome, and others, single-unit of a complex behavior like speech, we must understand studies in monkeys led to the first correlations between not only the properties of individual cells and pathways cognitive processes (such as perception, attention, and but also the network properties of functional circuits decision making) and patterns of firing of individual cells in the brain. While network properties arise from the in specific brain regions. This work changed the way properties of individual neurons in the network, they behavior was studied both in experimental animals and need not be explainable in terms of the behavior of in humans; the focus now was on the information pro- individual cells. Computational approaches are helpful cessing in the brain that leads to behavior. for characterizing the system as whole, for obtaining The need for greater anatomical knowledge led to a formal descriptions of what the system is capable of renaissance of neuroanatomy, evident in the develop- doing, and for determining how the interacting constit- ment of new techniques for tracing connections be- uent elements account for system properties. tween neurons by Sanford Palay at the NIH, Walle Nauta This review focuses on the topic of memory, but one at MIT, Matthew and Jennifer LaVail at Harvard, and aspect of cognitive neuroscience. We have not at- Max Cowan at Washington University. The search for tempted to document fully the remarkable progress that new neuroanatomical methods and the need to bridge has been achieved in our understanding of how the anatomy and function led to the application of neuro- nervous system learns and remembers. Rather, we fo- imaging techniques (positron emission tomography [PET] cus on two key components in the study of memory, as scanning and functional magnetic resonance imaging viewed through thework that the three of us have carried [MRI]) to cognitive problems. This major advance, pio- out with our colleagues during the past severaldecades. neered by Marcus Raichle and Michael Posner and by The first component is concerned with analyzing what Seiji Ogawa, Ken Kwong, and others, made it possible memory is, where it is stored, and what brain systems to relate changes in activity in large populations of neu- are involved. This is the systems problem of memory. rons to specific cognitive acts in living humans. By com- The second component of memory is concerned with paring the results of cellular recordings in nonhuman analyzing how memory is stored. This is the molecular primates and the results of neuroimaging in humans, it problem of memory. has become possible to study directly the neural correlates of sensory processing, motor actions, and cognitive processes. Where Are Memories Stored? In the 1960s and 1970s, there was also renewed inter- The question of where memory is stored emerged at est in the traditional discipline of neuropsychology. Early the beginning of the 19th century as part of the larger students of brain and behavior like Karl Lashley and question—to what degree can any mental process be Donald Hebb used the term neuropsychology broadly localized within the brain? The first person to address to encompass studies of experimental animals as well this question was Franz Joseph Gall, who made two as studies of humans. In this sense, cognitive neurosci- major conceptual contributions. First, Gall attempted to ence is the modern forum for the same topics and issues abolish mind–brain dualism. He argued, based on his that engaged Lashley and Hebb earlier in this century. Studies of patients with brain injury or disease that af- anatomical studies, that the brain is the organ of the mind. Second, he appreciated that the cerebral cortex fects mental function have always been a vital part of neuropsychology, and such studies formed one of the is not homogenous but contains distinctive centers that control specific mental functions. Gall therefore pro- foundations of cognitive neuroscience. As first clearly shown for language by Pierre Paul posed the idea of cortical localization. Gall asserted that Broca in 1863, patients with lesions of specific regions the brain does not act as a unitary organ but is divided of the brain exhibit quite specific cognitive deficits. Fol- into at least 27 faculties (others were added later), each lowing Broca and Wernicke, the neuropsychological at- corresponding to a specific mental faculty. He thought tempt at regional localization remained strong in Europe that even the most abstract and complex of human and in Canada but was in good part neglected in the traits, such as generosity and secretiveness, are localUnited States, with the exception of the work of Arthur ized to discrete areas of the brain. Benton, Hans-Lukas Teuber, and Norman Geschwind. Gall was not an experimentalist. He rejected the study As we shall see, continuing study of the behavioral con- of neurological lesions and the surgical manipulation of sequences of brain lesions proved to be a rich source experimental animals and instead attempted to locate

Review: Milner, Squire, and Kandel 447 Figure 1. Hebb and Penfield D. O. Hebb (right) and Wilder Penfield (left) in 1958 on the occasion of Hebb delivering the 24th Annual Hughlings Jackson lecture at the Montreal Neurological Institute. mental faculties by examining the surface of the skulls with the size of the cortical area removed, not with its of individuals well endowed with particular functions. specific location (Lashley, 1929). Many years later, with Perhaps not surprisingly, with this approach he misiden- additional experimental work, it was possible to arrive tified the function of most parts of the cortex. This ana- at a different understanding of Lashley’s famous contomically oriented approach to personality Gall called clusion. organology. Later, Gall’s associate, Gaspard Spurz- Perhaps the first effective answer to Lashley came heim, adopted the better-known term phrenology to de- from Donald Hebb (Figure 1, right). In his book The Orgascribe this approach. nization of Behavior, Hebb (1949) convinced many that Gall’s ideas were subjected to experimental analysis it was possible to think seriously about the brain proby Pierre Flourens in France in the late 1820s. Flourens cesses underlying memory. He developed concrete proattempted to isolate the contributions of different parts posals based on biological facts, taking into considerof the nervous system to behavior by removing from the ation the neuronal circuitry that might contribute to brains of experimental animals the functional centers memory storage. To explain Lashley’s result that learnidentified by Gall. From these experiments, Flourens ing could not be localized to a single brain region, Hebb concluded that individual sites in the brain are not suggested that assemblies of cells work together to sufficient for specific behaviors such as sexual behavior represent information and that these assemblies are and romantic love and that all regions of the brain— distributed over large areas of cortex. Sufficient numespecially the cerebral hemispheres of the forebrain— bers of interconnected cells will survive most lesions to participate in every mental function. He proposed that ensure that information can still be represented. The any part of the cerebral hemisphere is able to perform idea of a distributed memory store was far sighted. With all the functions of the hemisphere. Injury to a specific the accumulation of additional evidence, it has become area of the cerebral hemisphere should therefore affect apparent that no single memory center exists, and many all higher functions equally. parts of the nervous system participate in the represenDespite the findings of Broca and Wernicke on the tation of any single event. localization of language, the ensuing debate between Hebb influenced many students and colleagues—in cortical localization and equipotentiality in cognitive particular, Brenda Milner, who in 1957 described the function dominated thinking about mental processes, remarkable patient H. M. (Scoville and Milner, 1957). including memory, well into the first half of the twentieth H. M. had sustained a bilateral resection of the medial century. For example, in the period from 1920 to 1950, structures of the temporal lobe in 1953 to relieve severe this dispute could be followed in the work of Karl epilepsy. It was immediately evident following the surLashley, perhaps the dominant figure in American neu- gery that H. M. had a very profound impairment of recent ropsychology in the first half of this century. Lashley memory in the apparent absence of other intellectual explored the surface of the cerebral cortex in the rat, loss (Scoville, 1954). He could not remember what he systematically removing different cortical areas. In so had for breakfast, and he could not find his way around doing, he failed repeatedly to identify any particular the hospital or recognize members of the hospital staff brain region that was special to or necessary for the (except Scoville, whom he had known for many years). storage of memory. Based on these experiments, It seemed as though his life from the surgery onwards Lashley formulated the law of mass action, according was not contributing to his store of knowledge. He was to which the extent of the memory defect was correlated able to hold immediate impressions in his mind, but as

Neuron 448 soon as his attention was diverted they were lost. In to that of Penfield’s two patients, except that it was contrast, old memories from his childhood seemed to more severe. Again, there had been no intellectual loss; be intact. in fact, H. M.’s IQ had risen postoperatively, from 104 In fact, the encounter with H. M. was not the first to 117, presumably because he was having far fewer encounter with this kind of memory impairment. During seizures. His capacity for sustained attention was also the early 1950s, Wilder Penfield (Figure 1, left) began to remarkable. Thus, Milner showed that he could retain carry out unilateral removals of parts of the frontal or the number 584 for at least 15 minutes by continuous temporal lobe as a treatment for patients with localized rehearsal, combining and recombining the digits acinjury causing seizures. The temporal-lobe removals cording to an elaborate mnemonic scheme, but the motypically included the anterior temporal neocortex to- ment his attention was diverted by a new topic, the gether with the uncus, amygdala, and anterior parahip- whole event was forgotten. pocampal gyrus and hippocampus on the medial aspect H. M.’s success in remembering a three-digit number of the hemisphere. Milner and Penfield found that these for 15 minutes in the absence of distraction was at first removals produced at most mild material-specific mem- sight consistent with Drachman’s view that amnesics ory deficits that varied in kind with the side of the lesion. can hold a simple memorandum indefinitely provided But, unexpectedly, Milner and Penfield encounteredtwo that no interfering activity claims their attention (Drachpatients with a severe, persistent, and generalized im- man and Arbit, 1966). Yet it was already clear that for pairment of recent memory, following a removal limited H. M. verbal rehearsal played a key role in this holding to the left temporal lobe. Because both patients had process. In contrast, certain simple nonverbal stimuli undergone extensive preoperative testing, it was easy were forgotten by him within less than a minute. The to establish that this was a selective impairment of mem- evidence for this comes from delayed paired compariory, with no accompanying intellectual loss (Penfield son and delayed matching studies. and Milner, 1958). The impairment was manifested clini- In 1959, Konorski described a method for testing cally as a profound anterograde amnesia, such that the memory of single events, which was later adapted for experiences of daily life were forgotten as soon as the work with human subjects by Stepien and Sierpinski focus of attention shifted to a new topic. In addition, one (1960). This technique, called by Milner “delayed paired patient showed a retrograde amnesia covering salient comparison,” consists of presenting two stimuli in sucevents of the preceding few months and the other cession, separated by a short time interval. The subject showed a retrograde amnesia covering the 4 preceding must then indicate whether the second stimulus is the years. same as or different from the first. This means that subTo account for this unexpected memory loss, Milner jects must retain an impression of the first stimulus in and Penfield (1955) hypothesized that in each case there order to compare the second one with it. Task difficulty must have been a pre-existing, but undetected, atrophic may be increased by lengthening the intratrial interval lesion in the hippocampal region of the opposite hemi- or by introducing an intratrial distraction. Prisko (1963; sphere, so that when the surgeon removed the anterior cited by Milner, 1972) used the Konorski method to demhippocampus and parahippocampal gyrus in the left onstrate H. M.’s rapid forgetting of simple perceptual hemisphere, he effectively deprived the patients of me- material. She sampled five different sets of stimuli (three dial temporal-lobe function bilaterally. The reason that visual and two auditory), each set constituting a sepaMilner and Penfield focused on the hippocampal region rate task. The stimuli used were clicks, tones, shades was that one patient, P. B., had had his temporal lobec- of red, light flashes, and nonsense patterns. At least five tomy in two stages, and it was only after removal of the values were assigned to each variable, to prevent as far medial structures of the temporal lobe that the memory as possible the use of verbal mediation to bridge the loss was seen. Their hypothesis was confirmed 9 years retention interval. All paired stimuli were easily discrimilater, when P. B. died of a pulmonary embolism and the nable at zero intratrial delay. These proved to be exautopsy findings revealed thepresence of long-standing tremely easy tasks for normal subjects, who rarely made extensive right hippocampal atrophy, whereas the rest errors even with a 60-second delay and an interpolated of the right temporal lobe, including the amygdala and distraction. In contrast, H. M. performed all tasks well the parahippocampal gyrus, showed no significant ab- at zero delay, but with increasing intratrial intervals his normality. In contrast, on the operated (left) side, the 22 performance deteriorated sharply, so that at the 60- mm of the hippocampus that remained appeared to be second delay scores were approaching the chance level normal (Penfield and Mathieson, 1974). and were not further impaired by distraction. Milner and Penfield reported these two cases at the Subsequently, Sidman, Stoddard, and Mohr (1968) 1955 meeting of the American Neurological Association confirmed Prisko’s findings, using a delayed matchingin Chicago, and Scoville read their abstract. He called to-sample technique that allowed the plotting of discrimPenfield and said that he thought he had seen a similar ination gradients to show how far the subject’s choice memory disturbance in a patient of his (H. M.) in whom of a matching stimulus deviates from the sample stimuhe had carried out a bilateral medial temporal-lobe re- lus as the intratrial interval lengthens. In the nonverbal section, also in an attempt to control epileptic seizures. form of their task, H. M. was required to indicate which Penfield asked Milner if she would like to go down to one of eight ellipses matched the sample stimulus. With Hartford, Connecticut to study the patient, and that is zero delay he chose correctly most of the time, showing how the memory deficit in H. M. became more widely a normal discrimination of axis-ratios, but with increasknown. ing delays his performance deteriorated until, at 32 secClinically, H. M.’s memory disorder appeared identical onds, the sample no longer exerted any control over his

Review: Milner, Squire, and Kandel 449 Figure 2. H. M. Showed Improvement in a Task Involving Learning Skilled Movements In this test, he was taught to trace a line between the two outlines of a star, starting from the point S (Figure 2A), while viewing his hand and the star in a mirror. He showed steady improvement over the 3 days of testing, although he had no idea that he had ever done the task before. (The graph in Figure 2B plots the number of times, in each trial, that he strayed outside the boundaries as he drew the star.) Adapted from Milner (1962). choice. In contrast, H. M. had no difficulty with a verbal reaches the points of the star, one tends to move the version of the task, which required the matching of con- hand in the wrong direction. Eventually, with practice, sonant trigrams. However, as with other short-term ver- we can all learn to draw the outline of a star in a mirror. bal memory tasks, he succeeded only by constant re- It is a new sensorimotor skill, a visual-motor skill, and hearsal; his lips could be seen moving throughout the it is acquired across many trials. Milner was able to delay period. show that H. M. could learn that kind of task quite well. These and other related studies (Milner and Taylor, She took H. M. through 30 trials of mirror drawing spread 1972) concur in showing that H. M. can register percep- over 3 days, and he exhibited a typical learning curve (Figure 2B). Yet at the end he had no idea he had ever tual information normally, but that the information ceases to be available to him within about 30–40 sec- done the mirror drawing task before: this was learning onds. Milner (1972) suggested that such results support without any sense of familiarity. Nowadays, we are well aware that such dissociations are possible following a the distinction between a primary memory process with discrete brain lesion, but for Milner, looking at it for the a rapid decay and an overlapping secondary process first time, it was quite astonishing. Her finding contrib- (impaired in H. M.) by which the long-term storage of uted some of the early evidence that there is more than information is achieved. one memory system in the brain. Interestingly, even before the study of patient H. M. There Are Multiple Memory Systems in the Brain inaugurated empirical work on the different memory sys- H. M.’s failure on delayed matching and delayed com- tems of the brain, similar ideas had been proposed by parison tasks, which assess memory after a single pre- philosophers and psychologists on the basis of intuition sentation, did not rule out the possibility that he might and introspection. For example, in 1949, Gilbert Ryle, a be capable of some learning with intensive practice, or philosopher of mind at Oxford, proposed the existence indeed that certain kinds of learning might take place of two types of knowledge: knowing how, as in knowl- at a normal rate. Accordingly, Milner and her students edge of motor skills, and knowing that, as in the knowl- embarked on a variety of learning studies with H. M., edge of facts and events. Some years later Jerome including stylus maze tasks, both visual (Milner, 1965) Bruner, one of the founders of cognitive psychology, and tactual (Corkin, 1965). With one notable exception, called “knowing how” a memory without record. Memthese studies merely served to demonstrate H. M.’s ex- ory without record, Bruner argued, occurs in the case treme difficulties with new learning, as evident also in of experiences that “change the nature of the organism, his daily life. The exception was in the domain of motor change his skills, or change the rules by which he operskills, where, in 1962, Milner showed that H. M. could ates, but are virtually inaccessible in memory as specific learn a mirror-drawing task efficiently with stable reten- encounters.” Here, the neural machinery that supports tion from day to day (Figure 2A). a behavior is presumably modified directly. He called If one is shown a picture of a double-margin star “knowing that” a memory with record, a repository of (Figure 2) and asked to draw a line between the two information about the facts and events of everyday life. margins, one can do that very easily. However, if one has The demonstration of intact motor skill learning in to do it while seeing one’s hand and the star reflected in patient H. M. marked the beginning of a period of experia mirror, then it becomes quite difficult. When one mental work that eventually established the biological