Chapter 27 Chapter 16 Cognitive Psychologist-1

George A. Miller, although he looked quite young in 1960, although he was nearly 40 years old, he seemed to have a penchant for mischief. He was a professor in the Department of Psychology at Harvard University. His prestigious status and comfortable lifestyle are already guaranteed, and the future is really bright.And yet, this year, for all his uneasiness, he felt an irresistible urge to expose himself, even if it meant giving up his position at Harvard. His self-exposure was not motivated by radical political beliefs or imminent sexual affairs, both of which were commonplace at the time.His problem was his interest in thinking.

thinking?What's there to spoil or argue about?Isn't this the central question of psychological relations? No, it wasn't like that at the time.This has not been the case since behaviorism began to dominate American psychology 40 years ago.For the behaviorist, the invisible, immaterial, and speculative mind is an outdated metaphysical concept that no experimental psychologist concerned with his own future or reputation will touch. This topic, let alone spend time on this topic. But over the past few years, Miller has slowly become a converted spiritualist.Born and raised in Charleston, West Virginia, he had no interest in, or even a slight distaste for, psychology as a freshman; One psychology textbook has illustrations of the brain and other organs, "Because I grew up in a family of Christian scientists, I was taught to avoid drugs and to recognize a demon if I encountered one."

Either through education or obsession, he changed his view of the world.During his sophomore year at the University of Alabama, under the influence of a girl (whom he later married), he went to an informal psychology lecture.The lecturer was Professor Donald Ramsdell.Miller was so impressed with Professor Ramsdell that, a few years later, when he had completed his master's degree in language and communication, Ramsdell offered him a job teaching psychology undergraduates, although M. Le had never formally taken the class before.By this time, Miller was married and fathered, needed a job, and agreed.After teaching psychology for a year, he transformed himself.

He went to Harvard for graduate studies and got a solid basic education in behaviorist psychology.He made himself a distinguished student, and after his doctorate he became a lecturer.He spent the next 14 years, first at Harvard and then at MIT, where he conducted experimental research on language and communication.Despite his education, the work, unlike the rat-based experiments, forced him to think, whether he wanted to or not, about human memory and other higher mental processes.He came closer to idealism after attending a summer seminar at Stanford University where he worked closely with the linguistic psychologist Noam Chomsky.He spent a year on sabbatical at the Center for Advanced Study in the Behavioral Sciences in Palo Alto and acquired a number of new methods for the study of thinking, particularly stimulation of thought processes through computer programs.

Miller returned to Harvard in the fall of 1960, but he was a different man.As he stated in his memoirs: I realized that I was deeply dissatisfied with the narrow meaning of psychological concepts defined by the Harvard Department of Psychology.I just had a year of playing and playing in the sun.Returning to a world bounded by psychophysics on the one hand and operant conditioning on the other is a great reluctance.I decided that either Harvard would let me start something like the kind of exchange-inspired Stanford research center, or I would go. Miller told his friend and colleague Jerome Bruner about his discontent and his dream of establishing a new center dedicated to the study of mental processes.Bruner understood his feelings and saw his intentions.Together, the two of them went to Dean McGeorge Bandy, obtained his approval, and established the Harvard Center for Cognitive Research with funding from the Carnegie Corporation.Taking on such a name makes Miller feel like an open apostate:

To me, even as late as 1960, using the word "cognition" was an act of rebellion.Of course, it may not have been so serious for Jerry [Bruner]; social psychologists have never been wiped out by the popularity of behaviorism in the same way that experimental scientists have.Yet to a man who has accepted the science of reduction and should respect it, "cognitive psychology" is an unmistakable statement.It means that I am interested in thinking - that I have emerged from the shadows. He became the leader of a movement that dramatically changed the focus and research methods of psychology, and has guided the direction of psychology ever since.

George Miller stepped forward, a typical event in experimental psychology in the 1960s.A few, at first, and then most, abandoned rats, mazes, electric fences, and food-dispensing levers for the study of higher human mental processes.Throughout the 1960s, the movement gained so much momentum that it was dubbed the "Cognitive Revolution." Many factors contributed to the formation of this movement.Over the past 20 years, Gestalt psychologists, personality researchers, developmental psychologists, and social psychologists have all explored mental processes in different ways.Developments in other fields of science (some of which we have already heard about, and others we will learn about shortly) also happen to produce other facts about the workings of the mind.Specifically:

-- Neuroscientists who use microelectrode probes and other new technologies to observe neural phenomena and cellular exchange connections involved in thought processes. —Logicians and mathematicians developed information theory and used it to explain the capabilities and limitations of human communication. ——Anthropologists analyze the thinking patterns in different cultures. They discover which mental processes vary according to the nation, which are the same globally, and thus determine which are innate. —Psycholinguists who study language acquisition and use know that the mind acquires and controls the complex system of symbols we call language.

—Computer scientists, a new breed of hybridists (part mathematician, part logician, and part engineer) who propose whole new theoretical models of thinking and devise Some mechanical devices that seem to be able to think. By the end of the 1970s, cognitive psychology and these related fields were slowly forming various cognitive disciplines, or collectively referred to as cognitive science.Far from being merely a deepening and broadening of psychology on an unprecedented scale, the cognitive revolution is extraordinary—indeed, arguably utterly unbelievable—that it is the simultaneous development of six sciences of new knowledge of mental processes. develop.

Computer science had the greatest impact on psychology at the time.This new field of research was the result of intensive research during World War II, when the Allies desperately needed computing machines that could crunch huge numbers quickly to guide anti-aircraft artillery, operate navigational equipment, and more.However, even very high-speed computing machines require human operators to tell them what to do and what to do next after each calculation, which seriously hinders their speed and leads to inaccurate calculations.By the 1940s, mathematicians and engineers were giving machines a set of instructions (programs) stored in their electronic memory.Now, machines can quickly and accurately direct their own operations, follow through on long sequences of operations, and make some decisions about what to do next.Computing machines became computers.

In the beginning, computers only dealt with numbers.However, as mathematicians John von Neumann and Claude Shannon and other computer experts were quick to point out, any symbol can represent another kind of symbol.A number can represent a letter, a series of numbers can represent a word, and a mathematical calculation can represent a relationship expressed in language.For example, the = sign can represent "the same as...," the ≠ sign represents "different from...", and the > represents "greater than" or "too much." Set a set so that words become numbers and logical relations, or vice versa With the same rules, the computer can perform some operations similar to some human reasoning. The idea that computers might function as minds in some ways—at the time, the idea sounded more like science fiction than science—was first developed by von Neumann and neurophysiologist Warren McCulloch in 1948 at the California Institute of Technology. It was submitted for discussion at an academic conference on "Brain Mechanisms of Human Behavior" held by the University. The idea appealed to Herbert Simon, then a young professor of political science at Carnegie Institution (now Carnegie-Mellen University).However, "a professor who lectures on political science" barely describes him.Simon, the very bright son of an electrical engineer, skipped grades at school and was younger than his friends and classmates.He doesn't like sports, having grown up in Wisconsin, he's well aware of his Jewish background, so it's not surprising that he comforts himself by making himself an intellectual, but, in reality, his interests are diverse and broad Although he became a political scientist, he was interested in mathematics and taught himself mathematics, economics (for which he won the Nobel Prize in 1978), management, logic, psychology and computer science. In 1954, Simon and one of his extremely bright young graduate students, Alan Newell, discovered that they had a profound understanding of computers and thinking (both of whom would later earn degrees in psychology) and of creating a thinking computer program. big interest.In the beginning, they chose a very limited kind of thinking, namely the laws of verification in formal logic, which is an entirely symbolic and almost algebraic process.Simon's task is to ask for a proof of the axiom, "Not only to break it down as much as possible, not only to prove the steps, but also to find some hints to guide me." Then, the two of them tried to combine this information into a flow chart, and they put this A flowchart becomes a kind of computer program. A year and a half later, Simon and Newell stunned the attendees at an information theory symposium at the Massachusetts Institute of Technology in 1956.They described their intellectual product - "logic theorizer".Running for the first time on a primitive mainframe made of vacuum tubes, the Yoniac, the program was able to prove a series of axioms in logical form, each in less than 1-15 minutes. (With a modern computer, it may not take long enough to blink an eye.) "Logic Theorizer" was the first artificial intelligence program, and it was not very intelligent at the time.It can only prove axioms of logic—in about the time it takes an average college student—and must do so in algebraic notation.However, as the first computer program that can perform some thinking-like activities, it is indeed a ground-breaking achievement. By the end of the second year, in 1957, Newell and Simon, along with a college student, Clifford Shaw, had written another, much cleverer program, the GPS (General Problem Solver), which synthesized a A broad set of principles, similar to many intellectual tasks, including proving geometric axioms, solving cryptographic arithmetic problems, and playing chess. GPS will go one step ahead, or explore first, start to determine the "problem space" (the region containing all possible steps between the starting state and the desired goal), look at the results to make sure this step is closer to the goal, and adjust The next possible steps are tested to see which one brings it forward one step closer to the goal, and if the chain of reasoning strays from the direction, rewinding to the last decision point and starting over in another direction. A simple problem that GPS was able to solve easily in the early days was as follows (the problem is not expressed in these words, because GPS does not understand, but expressed in mathematical symbols): A fat father and two young sons must cross a swift river in the forest.They found an abandoned boat to paddle across, but would sink if overloaded.Each child weighs 100 lbs.The combined weight of the two children is equal to that of the father, and the boat can only carry up to 200 pounds.How do father and child cross the river? The answer, while simple, requires taking a step back in order to move forward.Two children cross the river in a boat, one goes ashore, and the other paddles back to ashore; the father paddles out and disembarks, and the other side of the child paddles back, pulls the other side of the child up and paddles together across the river. GPS is doing something similar to the human mind in designing and testing this solution.Through the same kind of heuristic process—extensive exploration and evaluation—it can solve similar but much harder problems. Two fundamental features of GPS (and later artificial intelligence programs) brought profound changes to cognitive psychology because they gave psychologists some more detailed and actionable conceptualizations of mental processes than they had previously It has everything, but also has a practical way to investigate the problem. One of the characteristics is representation: the use of symbols to represent other symbols or phenomena.In GPS, numbers can represent words or some relationship, while in the hardware that GPS operates on (i.e. the actual computer), groups of transistors represent these numbers by switching binary switches on and off.By analogy, cognitive psychologists can regard images, words, and other symbols stored in the mind as representatives of external phenomena, and the neural responses of the brain as representatives of these images, symbols, and thoughts.In other words, a representative corresponds to what it represents without being exactly like it.But this is actually old wine in new bottles; Descartes and Fermat discovered long ago that algebraic equations can be represented by lines in graphs. The second characteristic is information processing: the transformation and manipulation of data through programs to achieve a goal.In the case of GPS, the incoming information—that is, the feedback of each step—is evaluated in terms of where it leads, used to determine the next step, stored in memory, recalled when needed, and so on.By analogy, cognitive psychologists can think of the mind as an information-processing program that turns perceptions and other incoming data into mental representations and evaluates them step by step, using them to determine the next step on the way to a goal, Add them to memory and recall them when needed. Information processing (IP), or the "computable" model of thought, has been a guiding metaphor in cognitive psychology since the 1960s, and has enabled researchers and theorists to explore the inner universe in unprecedented ways. An example of such an exploratory approach illustrates how IP patterns allow cognitive psychologists to determine what is going on in the mind.In a 1967 experiment, a research group asked subjects to say aloud as quickly as possible whether two letters projected on a screen had the same or different names.when subjects see They said "same" almost immediately when they saw Aa, and they said "same" at about the same time when they saw Aa.However, the researchers used a high-precision timer to measure a very small difference.On average, subjects answered AA in 549 ms and Aa in 623 ms.It's certainly a subtle difference, but a statistically significant one.What explains this difference? The IP model views any simple cognitive process as a series of actions taken step by step in the form of data.The following simple flowchart is a typical diagram that cognitive psychologists often draw to symbolize what happens when we see and recognize things: This could explain the response-time difference in the experiment.If an image goes directly from the original "processing" box to the "awareness" box, it will go faster than if it has to go through the other two or three boxes to get there.In order to distinguish whether the letters contained in AA are the same letter, the subject only needs to complete the visual pattern discrimination in the visual image; in order to distinguish whether the letters in Aa are the same letter, the subject has to use the Locate the name of each letter and see if they're the same—74 milliseconds of extra processing, a small but inevitable difference, and evidence of how the mind accomplishes this small task.Thus, even a trivial experiment based on the IP model can show what's going on inside the mind. Rather, the finding is an inference drawn from the results rather than a direct observation of the process.However, unlike the behaviorist dogma, it is unreasonable in "hard" science to infer an unseen thing from an outcome.Geologists infer past events from sedimentary layers, cosmologists infer the formation and development of the universe from the ancient light of distant galaxies, physicists infer features from the traces of instantaneous atomic particles left on fog chambers or latex, Biologists have deduced the evolutionary path of human beings through fossils.The inner universe of the mind is explored according to the same methods: psychologists cannot go into it, but they can infer its workings from traces, so to speak, of the visible thought process. The explosion of cognitive neuroscience, also known as behaviorist neurobiology, has also shed a different light on cognitive processes.This specialized branch of biology seeks to explain mental processes at the level of neural phenomena; we have seen an example in the experiments of Huber and Wiesel, which historically found some retinal cells. Neuroscientific methods have been around for a long time, going back at least to the days of Descartes.Though he believed that thought was immaterial, he conjectured, as we have seen, that reflexes are caused by the flow of "vitality" through the nervous system, just as the movements of the automatics in the royal garden are impelled by the flow of water in the pipes. Memory is caused by the enlargement of special "brain pores" through which vitality can pass during learning.Likewise, a century ago, the young Freud confidently declared that all mental processes could be regarded as "quantitatively determinable states" of neurons, although he was quick to admit with chagrin that such The time is not yet ripe for understanding. The same hope continues to inspire many researchers.Over the past 40 years, and especially over the past 15 years, extraordinary advances in cognitive neuroscience have led some zealots to declare that concepts like needs, emotions, and thoughts are being replaced by biological data.A neuroscientist, Paul Churchland, declared that when humans have access to such data: We will proceed to reconsider internal states and activities within the framework of the real abundance that is finally here.Our explanations of how people behave with each other will take advantage of physiological states like neurodrugs, neural activity in specialized anatomical regions, and whatever state the new theory sees as relevant. Cognitive neuroscientists—some of whom are neurobiologists who have studied psychology and others who have studied neuroscience—do not work with thought processes, but with “wetware” These "wetware" are the trillion to two trillion neurons that make up the human brain.What intrigued them were phenomena like sodium and other ions that flow in and out as electrical impulses along the neuron's elevating shaft (stem pole); the molecular structure of the place); the neurotransmitter molecules that jump from one neuron to another neuron with excitation or inhibition information across the microscopic synaptic cleft; and the neural channels and neurons triggered by different stimuli and psychological activities The internet. Some of the work is also comical.One researcher implanted 16 microelectrodes into the muscles of a male grasshopper in order to record the electrical impulses from his neurons during copulation.Others inserted microelectrodes into the left foreleg of cockroaches and the feet of snails to measure nerve impulses as they crawled toward their targets.Investigators consider these studies of "motivated behaviour." Memory is fundamental in all cognitive processes, especially in higher species.Cognitive neuroscientists have been trying to figure out how and where memories are stored at the molecular level.Here are some ways they conduct their research: — As early as 1949, Donald Hebb, a Canadian psychologist, postulated that memory storage is formed by modifying the synapses that connect neurons (this is much the same idea as Descartes ).Repeated firing of a synapse during a learning experience somewhat strengthens the synapse and connects the two neurons into a kind of circuit, or "memory trace," he said.Hebb's hypothesis was somewhat borne out in 1973, when Timothy Bliss, a British neurophysiologist, and his colleague TJ Lomo measured the voltage in a neural channel in the brain of a rabbit, and then moved along The channel repeatedly delivered current, and it was later discovered that the channel was carrying a higher voltage than usual.Synapses have been strengthened by electrical impulses.The implication is that this is what happens when you learn. — Also in the early 1970s, an American psychologist, William Greenow, raised rats in two environments, one containing toys, mazes, and other stimulating devices, and the other completely empty of.Mice raised in a stimulating environment had more dendrites and thus made more synapses than those raised in a boring environment.Later, using electron microscopy, Greenouf and a colleague actually found that the brains of the more enriched mice had 20 to 25 percent more synapses in the affected cortical regions than the other mice. .Learning generates additional connections; memory traces must somehow be recorded in these synapses. --In recent years, Daniel L. Alken and colleagues at the National Institute of Neurological and Communication Disorders and Stroke have trained a species of marine snail to develop a response to light that it does not normally produce reaction.This marine snail quickly crawls towards the light source.Also, when the current is strong, it instinctively spreads out its tentacles to grab something on the surface.Alken combined these responses.They shone light into the pool and stirred the water, thus conditioned the snail—taught it—to deploy its tentacles whenever it saw a flicker of light.He then found that in some of the snail's photoreceptor neurons, molecules of PKC, a calcium-sensitive enzyme, were transferred from the interior of the neuron to the membrane, where they reduced the flow of potassium ions— This is a molecular-level explanation of memory. -- Over the past 30 years, James L. McGregor and other researchers have injected epinephrine (a hormone secreted by the adrenal glands) and other catechin-like neurotransmitters into mice that have learned to navigate mazes .Adrenaline specifically caused the mice to remember what they had learned longer than mice that had not been given the drug.One explanation, emerging from other studies, seems to be that a by-product of adrenaline overcomes opioids, a group of neurotransmitters that serve some useful purpose but clog the receptors on the receiving side of the synapse. receiver.The result: more receptors stay open, synapses function higher, and memories are strengthened. The neuroscientific approach to memory and other mental processes has great philosophical implications.It seems to have put an end to the ancient topic of the soul and the body once and for all, because it can explain all mental processes in terms of matter and phenomena.Memories and other higher mental processes are just ions and molecules flowing in the brain's vagus nerves and tiny channels. However, most cognitive psychologists agree that neural phenomena do not provide adequate or useful explanations for cognitive phenomena.Few are dualists who believe in the sense of immaterial thought, however, emphasizing that mental processes, though composed of neural phenomena, are properties of the organization or polyarchy of these components, not of the components themselves, as if to say Now, shelter is not a property of masonry, beams, and slabs, but of the house made of these things. The Nobel laureate Roger Sperry, himself a brain scientist, offers another analogy: Higher mental processes are like a wheel rolling down a hill, with the roll determined by the wheel's "overall system properties" rather than being determined by atoms and the molecules they are made of. The developmental psychologist Jerome Cagan used another analogy: the beautiful laws of planetary motion reveal phenomena that cannot be described by the atoms they are made of. Another analogy is that of the cognitive scientist Earl Hunter: "We know from physiological measurements that the left temple area of ​​the brain is active when we read, but we cannot be sure that this activation is caused by reading Shakespeare." The writings, or by reading the works of Agatha Christie triggered." Finally, there is a quote from the cognitive psychologist George Mandler: "The mind has functions that are different from those in the central nervous system, just as society functions in ways that cannot be understood from a single It is the same as deduced from some functions derived in the mind." Most cognitive psychologists therefore believe that a single word drawn from memory is not comparable to the millions of firings of neurons and their resulting millions or even billions of synaptic transmissions. Rather, it is the product of these structures or patterns of emission or transmission.Neurophysiological research on memory is valuable, but it cannot tell us how to learn something, how to recognize something we have experienced earlier, or retrieve something from memory-such as how to speak when we speak. some words used.These phenomena, or epiphenomena, are grasped not by cognitive neuroscience, but by cognitive psychology. The distinction between basic phenomena and epiphenomena can be demonstrated by the previous title of "Cognition" magazine: The pattern is made of ink molecules on paper, something that has nothing to do with its meaning.At a higher level of organization, molecules make up letters, which themselves are meaningless symbols, but here organize to form the word "cognition."However, we are not done yet.Although this design looks real, but three-dimensional, it is something that cannot exist in the real world; this illusion of plausibility is a psychological epiphenomenon.If possible, explain it in terms of ink, molecules of letters, or as energy firings from neurons in the visual cortex. In psychology itself, at least in academia, the cognitive revolution quickly won the approval of some senior psychologists as well as the enthusiasm of most young psychologists and most psychology graduate students.At first, like neuroscientists, they focused on perception first, but quickly shifted their attention to the use of perception by the mind—its higher mental processes.By 1980, the theorist of mental processes John Anderson defined cognitive psychology as "an attempt to understand the nature of human intelligence and how people think." According to information processing theory, the most basic first step is the storage of data that enters memory, whether for a fraction of a second or for a lifetime.As James McGough put it in a recent lecture: Memory is essential to behavior.There is nothing of importance that is not fundamentally based on memory.Our consciousness and our actions are made of experience.And, our experiences make us who we are because they have long-term impact. How vital memory is to thinking is painful and vivid for anyone who knows someone suffering from the profound effects of Alzheimer's disease.He might forget what he wanted to say in the middle of a sentence, might suddenly get lost walking down the path to his mailbox, not recognize his own children, be suddenly unfamiliar with his own message, and throw a fit of rage. George Miller gave a speech at a meeting of the East Coast Psychological Association in 1955 that became a landmark for cognitive psychology theorists who study memory.Miller, with his usual snappy wit, called the speech "the mysterious seventh, plus or minus two."He began by saying, "My problem is that I have been tormented by a one-digit integer." The integer is 7, and what Miller finds so mysterious and unbearable is that, as many experiments have shown, one can It is often this number that is immediately transferred to memory. (People can temporarily remember one-digit numbers like 9237314 after a short period of study, but not numbers like 5741179263.) It is remarkable, and mysterious, how short the limiting factor in what we can notice in temporal memory can be.Confinement can serve a crucial function: it greatly tailors the incoming data into something the mind can urgently need at any time to notice and make decisions, a function that undoubtedly helped our primitive ancestors in the jungles and deserts. survival.However, it also raises some puzzling questions.How does such a small attentional area handle the flood of perception that we have to pay attention to while driving or skiing?Or, in a person talking to us.Or how sound and meaning get mixed up when we try to say something to them? However, even with program chunking, the energy of temporary memory is still insignificant compared with the vast amount of material—our daily experience, language, and general information of all kinds—that we learn and store in long-term memory. something in memory and recalled when needed. To explain this inconsistency, and to determine how memory works, cognitive psychologists conducted many experiments in the 1960s and 1970s, the results of which strung together to form a complete picture of the information processing of human memory.In this picture, memory is made up of three forms of storage ranging from fractions of a second to a lifetime.Experience or information items that only need to be used for a few seconds fade away soon after use, and may be transformed into semi-permanent or long-term long-term memory.Researchers and theorists describe the type and transmission of information in a form similar to the flow chart that follows. The simplest form of memory consists of sensory "buffers" where incoming sensations are first received and retained.The researchers demonstrated by spinning solid mirrors that the cache exists, and they also measured how long long-term memories remained in the cache before disappearing.In a classic experiment from 1960, psychologist George Sperling flashed the following letter pattern on a screen and asked his subjects to observe carefully: RBLA QUR GKRX The letters flashed for about one-twentieth of a second, too short a time for the subjects to see all the letters, although they were able to write down any line of letters immediately after seeing them. (After the flash, a voice tells them which line to write.) They can still "see" all three lines of letters when they hear the voice, but by the time they finish writing one line, they don't know the other lines .The memory has disappeared in less than a second. (Some experiments with sound by others have yielded comparable results.) Apparently, incoming perceptions are stored in a buffer and disappear from there quickly—fortunately, if they remain longer , the world we see will be a blur. However, since we need to keep what we care about for a longer period of time, another form of temporary memory that lasts longer is necessary.When we pay attention to the material inside the sensory bumper, we can do so in any of several ways.A number will not only be a perceived shape, but a symbol - 4 will have a name (four), and it will have a meaning (the quantity it represents); similarly, we read or hear words will have meaning.This process transfers what we're paying attention to from the cache into what Miller is talking about in short-term and immediate memory. 按一般人的话来说，短期记忆是指最近几小时或者最近几天的事件的保留，可是，按行话来说，它是指任何成为当前心理活动中的一部分，但在使用过后不再保留下来的东西。这种形式的记忆是短暂的。我们所有人都有这样的经历，找到一个电话号码，然后遇上占线，得重新再找到号码才能拨号。但是，我们可以自己把这个号码念几次后使其在脑海里保持好多秒甚至好几分钟——心理学家把这种活动叫做“预演”——直到用完为止。 因此，为了测量短期记忆的正常保留期，研究者们只好防止预演。印第安那大学的一组研究人员1959就进行了这样的实验。他们告诉受试者说，他们得努力记住三个三个一组的辅音，这很容易做，可是，他们一看到辅音时，得根据节拍器的节奏倒着念辅音，这把他们事先的注意力倒空了，使预演不可能产生。研究者在不同的时间内使受试者倒读的活动突然停止，看看他们能够把三个辅音保持多长时间；没有一位受试者的保留时间超过18秒的。许多后来的实验都确证，短期记忆力的衰退时间为15－30秒钟之间。 最近，其它研究在两种短期记忆之间作了区分（上图并未表现出来）。一种是语言的：我们刚刚讨论过的对数字、单词等的直接记忆。第二种是概念性的：一个概念或者通过一个句子或者其它好几个部分的表达传达出来的意义（比如某个代数方程）。在1982年进行的一项实验中，让受试者看一些句子，每次看一个单词，每个单词只给十分之一秒的时间；他们可以轻松地记住如下所示的有效句子（不过不一定是真实的）： 愚蠢的学生讨厌没有经验的老师。 但对于同样长度的无效句子，他们就束手无策了，比如： 紫色具体培训想象性的苍道。 更早一些时候（1967年）进行的一项实验已很清楚地显示出，我们很容易在短期记忆中保持住一个句子的意义，可是，很快会忘掉一些准确的词。研究者雅克琳·S·萨奇让受试者读下面这句话： 在荷兰，有一位名叫利普谢的男人专做镜片。有一天，他的孩子们在玩一些镜片，他们发现，如果两片镜片放在约一英尺远的间隔内时，东西好像更近些。利普谢开始实验，接着，他做的“间谍镜”吸起了人们极大的注意。他就此给伽俐略写了一封信，就是意大利那位伟大的科学家。伽俐略建起了自己的仪器，第一个晴朗的夜晚就把仪器抬了出来，他很惊奇地发现，空阔的黑色夜空里挂满了星星！ 接着，萨奇问他们，下面这三个句子里，哪一句出现在他们刚刚看过的原文中： 1.伽俐略这位伟大的意大利科学家，就此给他寄了一封信。 2.他给伽俐略这位伟大的意大利科学家就此寄了一封信。 3.他就此给伽例略写了一封信，就是意大利那位伟大的科学家。 结果是：萨奇的受试者都知道第一句不在原文中，可大多数人都说第二句在原文中。They are wrong.它的内容当然是正确的，可其形式却不对，他们在读第二句之前读过第一句，而衰退时间长得足以让他们忘记用词和内容。 同样的，我们可以在“长时间记忆”中，将我们谈过的一些话，我们读过的一些书，我们经历过的一些事情的要点和我们得知的无数的事实保持几个月，几年或者终生不忘，可是，没有人，或者只有很少一些人，能够记住这些事情发生时准确的用词。以这种方式记忆住的大量的材料比我们大多数人能够记忆住的多得多：数学家约翰·格里菲斯计算过，一般人类的记忆终生的记忆能量是十的十一次方（100个百万的立方）比特（在信息学说中，比特是最小的信息单位，它等于一个简单的是或者不是。一个数字或者字母表中的一个字母等于好几个比特），或者是《大不列颠百科全书》里面所包含信息的500倍。 短期记忆中的新信息在我们使用之后就遗忘了，除非我们使其经过进一步的处理变为长期记忆的一部分。一种形式的处理是死记硬背，如小学生背诵乘法表一样。另一种是将新信息与某些很容易记忆的结构或者助记忆术联系起来，如单调的儿歌（学龄前儿童背诵字母表的歌）或者是押韵规律（“看见字母C，记得加上E”）。 可是，最为重要的是在60年代和70年代进行的实验中变得非常明显起来的一种，即“精细处理”，根据这种方法，新信息与我们现存的有组织的长期记忆联系起来。我们可以说是将它叠接到语义网中了。如果新项目是一枚我们以前从未见过的芒果，我们会把这个词和概念与合适的长效记忆联系起来（不是一种物理的位置——思想和图象现在被认为是散布于大脑中的——而是一种概念上的位置：即“果子”这个范畴），连同芒果的视觉图象、触觉、口味和嗅觉（我们将它们分别列入图象、触觉品质等等的范畴中），再加上我们所知的、有关它的生长地、它的价格等等的其它信息。将来，当我们试图想起芒果时，我们会以上述任何一种办法检索出芒果来：通过回忆它的名字，或者思考水果，或者回忆有青皮的水果，或者想起黄色的切片，或者任何其它的范畴或者联系起来的特征。 所有这些信息是如何组织起来的，对此，我们已知的许多东西都是通过反应——时间实验得知的，比如请受试者在很短时间内说出尽量很多红色东西的名字，或者说出水果的名字，或者说出一些以某个字母开始的东西的名字。利用这些方法，华盛顿大学的伊丽莎白·洛夫特斯发现，在一分钟的时间内，志愿者平均可以说出12种“鸟类”，但“黄颜色的”东西一分钟只能说出9种来。她的结论是，我们不能在记忆中直接地找到符合某种特征的东西，但能很快地找到范畴（鸟、水果、蔬菜），并在每个范畴里寻找到这些东西。 同样地，如洛夫塔斯和另外一个同事阿伦·柯林斯所发现的，人们面对“鸵鸟是一种鸟”这样的说法，回答“是”或者“不是”，比面对“金丝鸟是一种鸟”这样的说法所花的时间要长些。含义在于：金丝鸟是一种比鸵鸟更加典型的鸟类，它更接近于范畴的中心，因此需要较少的时间就可以辨认出来。 到1975年，柯林斯和洛夫塔斯在这样一些资料的基础上，象征性地把长期语意记忆描述成复杂的网络，它是层次性的（一个总的范畴围着具体的例子），也是联想性的（每个例子都与一种特征相联）。他们用下面这个图描述了这种情况： 这还只是语义记忆网络中微不足道的一个例子。图中的每个结点还与其它许多的结点链连接在一起，这里没有显示出来，比如“游泳”也许可以跟“鲸鱼”、“人类游泳运动员”、“运动”。“有益的锻炼”连接在一起，而这每个词又可以与其它许多的词和特点及其它东西连接起来，无穷无尽。 记忆研究已经伸展到了很远的地方，我们只得就此打住，看看其它许多发现的简短报告，然后继续往前走。 概念化：许多研究证实，人类思维具有一种倾向，会自动地把一些类似的物体在记忆里面组成相同类别，并从其相似性中找出总体的概念或者范畴。哪怕只有几个月大的婴儿好像都知道进行简单地概念化工作。一项研究显示4个月大的婴儿会把不同的蓝色、绿色、黄色和红色分成类别。看到过一种色彩组的不同物体以后，它会显示出对别的任何色彩组的爱好。结论是：色彩分类要么是天生的，要么是在出生后迅速形成的。 许多其它的研究也曾统计过，当孩子获取语言能力时，他们会在有过与狗、猫、松鼠和其它动物的经验后，慢慢地发育像“动物”这类范畴的东西。确切地说，父母也向孩子们教授这种概念，可是，有一部分好像是天生的。这种倾向在所有民族中都如此常见，以致于成了一种天生的人类特征。人类学家布伦特·伯林发现，在12种不同的原始社会的民族中，人们都将植物和动物以惊人地相同的方式归结成类了，也就是说是以层次的方式进行的，从与生物学种类相类似的子类开始，并把一些类似生物学的种属的东西放在一个较大的门类之下，进而按照生物学中的植物和动物界的方式将一些范畴归结在一起。 概念化的能力也许是由进化选择而成的。它具有生存价值，因为从这些分组开始，我们可以对一些陌生的事物进行有效的推论。在最近的一项研究中，洛切尔·格尔曼及其一位同事让受试者看红鹤、蝙蝠和一只山鸟的照片。受试者被告知说，红鹤“这种鸟的心脏只在右边有一个动脉弧”，然后又对受试者们说，蝙蝠的心脏“只在左边有一个动脉弧”。然后，问他们说，山鸟“这种鸟的心脏有什么”？几乎百分之九十的人都正确回答说“只在右边有一个动脉弧”。他们的答案不仅以蝙蝠和山鸟的相似性为基础，而且以红鹤和山鸟的范畴为基础。哪怕4岁的小孩子面对这样一种类似和简单的测验时，百分之七十的时间里也会以范畴的成员关系为基础回答问题。 代表：研究者们一直就不了解材料存储在记忆中的形式。有些人相信，它是以形象和词汇双重形式存储下来的，这两种数据库之间还存在着交流。其他一些人则以信息理论和计算机模式为基础，认为信息只以“命题”的形式存储在记忆中。一道命题是一种简单的“思想单元”或者是一些像蝙蝠和翅膀（蝙蝠有翅膀）或者蝙蝠与哺乳动物（蝙蝠是哺乳动物之一）之间的概念关系的形式象征出来的知识。 在第一种看法中，蝙蝠会以一种图象的形式在记忆中与关系到自己的语言说明的形式记录下来；在第二种观点中，蝙蝠只会以一种关系的形式记录下来（如在图39中的语义网络关系中一样），虽然它不是语言的，但与“蝙蝠有翅膀”，“蝙蝠有皮毛”等等是相等的。另一个命题性的例子可以在下面的句子中看出来： 公主吻了青蛙。 及其被动态： 青蛙被公主吻了一下。 这两个句子表达的是同一个意思；它们都是语言的表达，焦点不一样，但它们是同一个命题，或者是同一个关系知识的单元。 每种观点的倡导者都有很多证据来证明自己的观点。我们早先看到罗杰·谢帕特做的“心理旋转”实验，就指明，我们看到物体时是以“心灵之眼”进行的，而且在对待这些图象时，就好像它们是三维的物体。在类似的一项实验中，阿兰·帕维沃问受试者一座钟上面的指针之间的角度是6：15还是11：25，是6：15时大些还是5：15时大些。他们回答说第一部分，因为这部分的角度差别大些，也更明显，对第二个问题的回答也快些。当受试者看着实际的钟时也是同样的情况。帕维沃合理的结论是，不看实际的钟就解决了问题的受试者是在心理之眼中“看着”这些钟的。 可是，命题论的提倡者也有同样好的证据来支持他们的观点。他们辩论说，图象不能传达像“有”，“引起”和“与……押韵”等的关系，也不能代表范畴和抽象的概念。赫伯特·西蒙和威廉·蔡斯发现，国际象棋大师只需几秒钟扫一眼棋盘就可以重复整盘棋的布局——但他们能够这样做的前提是，这盘棋必须是实际比赛中下的真棋。如果是随意摆的棋，棋子摆在任意位置上，他们就记不住了。含义是：大师的记忆不是视觉上的，而是基于几何关系——棋子的攻防移动位置。最后，计算机程序中的信息是以命题形式存储起来的，如果可计算性是一种好的认知模式，思维以同样的方式存储信息就不无道理了。 （还有第三种提法，是有些理论家赞成的，即，有好几种类型的心理代表方式：命题式的、心理模式型的、图象式的，每种方式都能在不同的抽象程度上对信息进行编码。） 对于这一点，最后的结论尚不得而知。 概要：1932年，英国心理学家弗雷德里·巴特利特给受试者讲一些非西方文化来源的民间故事，然后让他们复述故事。他们不能准确地记住故事，偶尔补充一些细节，修改一些事件，以便给发生的事件提供一些道理，并漏掉一些细节，因为这些东西对西方头脑来说没有意义。巴特利特的结论是：“记忆不是对无数固定的、无生气的和零碎痕迹的重新激发”，而是以我们自己有组织的经验体为基础的“想象性的重构，或者建筑”。他把这种组织起来的东西称作“概要”。 巴特利特的思想最近几年得到复苏。概要——也称作“框架”——现在被认为是对不同话题整合信息的包装，它们保持在记忆中，我们依靠他们来解释暗指和零散性的信息，一般的对话——甚至最有叙述性的写作是由它们构成的。1978年，当时还在圣迭哥的加利福尼亚大学的大卫·鲁美尔哈特报告了一些实验，他在这样的一些实验中给受试者读故事，一句一句地读，看他们是怎样以及何时形成对整个故事的清晰思想的。比如，当他们听到：“我被带到一间白色的大房子里，我的眼睛因为明晃晃的光线的刺激而眨巴起来”时，约有百分之八十的受试者立即猜想到，他们听的一定是在某个医院或者审讯室里的情景，并对他们听到的几个词提供了大量信息。如果下一句或者下两句与他们的猜想不一致，他们会把故事改过来，根据不同的概要重编故事。 最近进行的其它许多工作都确切地证明，我们理解并解释——或者说经常错误地解释——我们听到，看到和体验到的东西，是通过唤起我们的预期和有组织的知识结构来进行的。总起来说，记忆不仅仅是需要是可以唤醒的一种信息登记册，而且是能指导我们思维的程序。 遗忘：许多研究曾探索过，我们为什么会忘掉某些事情，但不会忘掉另外一些事情；怎样才能改善记忆力，特别老年人的记忆力，因为这些人中的大部分人都会经历某种程度的非病理性记忆力减损。（与年龄相关的正常的记忆力问题经常可以通过助记术和其它培训得以改善。还有一种可能是，就在不远的未来，人们会找到一种药物生理学的治疗办法，以重新平衡被更改过的神经传递器输出。） 一些最有兴趣的工作不仅与特定记忆的整体丢失有关，而且与重要细节及其由新材料所替代的遗忘有关。我们的法定系统在很大程度上依赖于一种假定，即如果我们记得某个事件，我们所记忆的一定就是事件的原委。法庭和许多心理治疗者也相信，遗忘的材料如果通过催眠加以检索出来，也一定就是实际发生的事件的真实记录。可是，心理治疗者早就有了证据，知道我们会修改记忆，以便使回忆出来的东西更易为自我接受。伊丽莎白·洛夫特斯也收集到了大量证据，显示一种令人震惊或者创伤性的事件会被创伤本身所扭曲，对一个事件的记忆会在一位有经验的检察官提出的有圈套的问题面前发生倾斜，随着时间的推移，我们会给记忆里增加新的信息，没有办法得出原来的真实情况。催眠有时候会检索出深埋于记忆深处的东西——有时候又会调出一些人为的东西来。 然而，我们几乎所有的人都确切地相信，某些事件永远地，准确地留在了我们的记之中，挥之不去，磨之不掉。对有些经验的回忆，如听到肯尼迪总统被刺的消息时，或者听说挑战者号太空飞船爆炸等，在心理学家们看来是“闪光灯泡记忆”，因为它们是非常生动的人生定格，很难忘掉。最近，艾莫利大学的阿尔里克·莱塞及助手尼可勒·哈尔什抓住了一次特别的机会，研究了这种现象。挑战者号空难发生（1986年1月28日）的次日，他们请一大批大学生记录了他们如何听说空难的消息。两年半以后，再请还能找到的这些回答者填写一份有关该事件的问卷，6个月后又加以采访。 有三分之一强的学生对该事件的时间、地点、谁告诉他们的，等等的回忆是完全错误的，他们在1986年的报告中的回答也是如此，另有四分之一的人有部分错误。当受试者看到他们自己原来的说法时，哈尔什和莱塞报告说：“许多人因为原来与现在的说法之间出现的差距而不安……有趣的是，许多人继续认为自己现在的说法是正确的，原来的说法可能有误。”错误是从哪里来的？哈尔什和莱塞把它们称作“叙述重构”，与巴特利特在1932年描述的类型一样。 有时候，哪怕是在快速发展的认知革命中，会发生更多此类变化…… 科学家从标本、事件、自然现象和这种或者那种实验发现中推论出自然法则。对认知科学家而言，可比的原始材料就是思想，可是，神经的排放，或者可以指示思想的脑波尽管可以通过示波图加以追踪，但却不能透露里面的任何东西出来。体态、表情、数学或艺术符号及演示（如在运动培训中）可以传达思想，但也只在非常狭窄的范围内。思维可以观察到的主要形式还是语言，因此，它也就堂而皇之地被称作“心灵的窗口”。 人们当然也可以说语言是思维的足迹，因为语言不仅仅传递思想，而且还在其结构中带有思维如何工作的痕迹。对通过这些痕迹显露出来的思想过程进行的研究是心理语言学家的地盘。（语言学是一门旧学科，它主要处理语言本身的特性。） 这里有一个例子，可以说明语言的痕迹：小孩子们倾向于把不规则动词和名词当作规则动词和规则名词进行处理（“小狗跑溜走了，”“那个小孩子长生了两粒牙”）。可是，他们并没有听到过成人这么说，因此也不是模仿所致。心理语言学家说，这种错误显示，孩子们能够辨认成人语音中的一些规则，如加上“ed”就可以形成一个简单的过去时，加上“s”或者“es”就可以做成名词，然后认为这种规则适用于所有动词和名词（这种倾向称作“过高归纳”）——这证明，人类思维自发地根据例子而形成概念，再把规则应用到新的情形中。 这只是心理语言学家们在语言中找到的思维过程留下的少数痕迹之一。这并不是只在英语中才有的情况。类似的情形可以在任何语言中找到，而且好像还是人类思维的特征之一。“我们对人类语言观察得越多，”洛衫矶的加利福尼亚大学领头的心理语言学家维多利亚·弗罗姆肯说，“它们越像是由同一些万有原则和制约所掌握着。” 这种万有原则当然不包括语法和词汇在内；从这个角度来看，英语、斯瓦希里语、巴斯克语可以说没有任何共同之处。可是，在听说这些语言而长大的孩子们中间，他们能够在不需要教授的情况下区别名词的单复数形式，代表过去和现在的动词形式等等，并为自己建造一套主管这门语言的规则。同样的，他们直觉地学会了掌握词序的基本规则，并能利用正常的词序建造一些简单的感叹句来。没有哪一个讲英语的孩子会说：“牛奶更多一些想我。”也没有哪个讲另外一种语言的孩子会把基本的词序搞错。 本世纪中叶以前，心理学与语言学几乎不搭界，可是，随着认知革命的到来，有些认知心理学家和语言学家看到了各自学科的新发展，并呼吁通过另外一门学科来扩张自己这门学科。比如，语言学中有关语法工作机制的某些新理论意味着，思维在处理概念时会执行一些行为主义心理学不能解释的复杂操作。1953年，一系列心理学家和语言学家在康奈尔大学举行了一次学术会议，讨论了它们共同感兴趣的领域，并采纳了“心理语言学”这个名字，以确定为语言心理学研究的名称。 心理语言学当时还是一个不太为人知道的新学科，4年以后，哈佛教授协会的一位29岁的年轻会员发表了一篇专题论文，从而使这门学问受到人们的注意。在这篇专题论文中提出来的理论，已经成为当前这个时代里心理学的两大重要发展之一（另一个是人工智能）。其作者便是诺姆·乔姆斯基，他的一些观点我们在前面已经听说过。 乔姆斯基有一头蓬松的头发，戴着眼镜，是位浑身皱巴巴的天才——可说是知识分子的典型。他差点就没有当成心理语言学家。他在大萧条时代纽约激进的犹太社区里长大，但他父亲却是一位知名的犹太学者。年轻的乔姆斯基还在小时候就已经掌握了闪族语的一些基本结构知识，也了解到语言学大致是什么东西。这两样，一样是缴进的政治学，一样是语言学，从此以后就主宰了他的一生，可有一阵子，乔姆斯基政治的一面几乎就克服了语言学的一面。他在宾夕法尼亚大学读过两年书，发现学术生活比较起左派政治生活来说十分枯燥无聊。1948年，他准备去以色列，准备在那里从事一种激进的、理想主义的阿拉伯-犹太工人阶级运动，这时，他遇到了泽尔格·哈里斯这位宾夕法尼亚大学的语言学教授。他们因为共同的政治信仰而相遇，可是，乔姆斯基很喜欢，也很敬佩的哈里斯让他对语言学产生了浓厚的兴趣，使他放弃了去以色列的打算，转而投身于语言学和左激政治运动。 当乔姆斯基遇见他的时候，哈里斯正打算发展一个基于行为主义原理的语言学系统，这个系统将能够解释语言模式而不必推论其意义。可是，他的计划有误，而且，在许多年里，乔姆斯基花了很多精力试图使其产生作用。当他达不到目的时，他放弃了哈里斯的理论，并在两年时间内形成了自己的理论。可笑的是，乔姆斯基是位左派分子，而他的学说的中心议题，如他在《句法结构》专题论文中所述的，却是说，语言知识和能力的某些方面是天生的，不是后天学成的，这个观点是左派分子、自由主义者和接受过行为主义培训的心理学家们认为是唯心灵主义和反动的。 乔姆斯基认为，孩子使听到的语言产生意义，以及获取语言的途径，不是通过语言的语法而来的，（“表层语法”，按他的说法），而是通过天生的能力来辨识听到的句子中间所包含的元素短语之间的深层句法关系而来的，这就是他称作起支撑作用的连接的“深层结构”的东西。他指出，作为一种证据，孩子们拥有一种轻松感，他们可以轻易地理解一种形式的句子转换成另一种形式的句子时的真实意义，比如，当一个陈述句转变成问句时，而且还可以自己完成这样一些转换关系。如果表层语法是孩子们所依靠的东西的话，他们可能会从转换句子中得出不正确的抽象。下面一些例子可以说明这些问题： 这个人很高。 这个人高吗？ 他们会得出一个规则：从开始处着手，转到下一个出现“is（是）”的地方，或者另外一个动词，再把动词移到前面去。可是，这个规则太简单了；遇到象下面这个句子时，他们就不起作用了： 个子很高的那个人在房间里—— 按理，他们会把这个句子转换成： 那个高人会是在房间里？ 可是，孩子们不会犯这样的错误。他们只犯一些很小的错误，如“牙奇”，而不会犯严重的错误。他们可以感觉到思维元素之间的关系——其句法构成或者“短语结构”。正是通过这种“万用语法”的知识方法，孩子们才使自己听到的东西产生意义，并毫不费力地构造自己从没有听说过的正确句子。 孩子们是在什么时候以及以怎样的方法来获取这种万用语法和深层结构的知识的？乔姆斯基的答案完美地代表了针对行为主义理论的一场革命，因为行为主义认为新生儿是一块白板。他认为，在大脑的某个地方，有一个专门化的神经结构——他把这个地方称作语言获得器，或者叫L． A． D——这个地方是靠基因连接起来的，可以辨认由名词词组和动词词组所代表的一些事物和动作与彼此作为主动者、动作和客体之间发生联系的各种方法。 乔姆斯基和其他采纳了他的观点，或者开发出自己的观点的一些语言心理学家们开始以新的形式回答一些老问题。这在行为主义时代是不允许进行的，这些问题涉及知识是否在经验之间就存在于思维之中。他们的答案是：虽然语言本身是学习得来的，但大脑的结构殊同一般，孩子们可以自发地从他们听到的东西里面抽取语言的规则，而不需要人们来告诉他们这些规则，虽然他们会犯一些枝节性的错误，但能在构造自己的句子时利用这些规则。 尽管乔姆斯基平常很严肃认真，但他也会有很幽默的时候。为了演示一个句子元素之间的深层关系，他构造了一个完全荒诞不经的句子，从此以后变得非常出名：“没有色彩的绿色思想愤怒地睡着了。”尽管这个句子毫无意义，可它对读者来说还是与这样一句同样没有道理的句子很不同：“思想愤怒地绿色毫无色彩睡着了。”任何熟悉英语的人都会觉得，第一个句子多少让人能够忍受一些——它几乎是能够表达什么事情的——而第二个句子完全是令人不快的垃圾堆。其理由是，第一个句子遵循了表层语法和深层语法的规则，而第二个句子却没有。 乔姆斯基引发了激烈的争议，很大一部分原因是因为他的先天论思想，尽管他并没有断定天生的思想，而只是说天生能够以有用的方式体验语言的能力。有些批评家排斥L． A． D．的假设后同意说获取语言的能力的确是天生的，但他们说，那是总体的知识能力的副产品。其他觉得天生的语言获取器很难接受的一些人也在不停地寻找证据，以期对之加以反驳。最近的一位是心理语言学家菲利普·利伯曼。他说，基因传递的器官服从变异是生物学的原则。果真如此，有些孩子就具有不正常的语言获取器，因而在语言理解的某些领域里就会出现不足情形，可目前好像还没有这方面的证据。 除开这些争议之外，在三十多年时间内，心理语言学家和认知心理学家一直在收集证据，以期显示语言与思维是如何发生联系的，并希望显示出思想过程。有些人很有耐心地观察孩子在学习语言的过程中出现的错误和自我纠正行为。有些人分析语言游戏，还有一些人研究发育语言障碍，如失语症和由大脑损伤造成的获取语言的丢失疾病。更有一些人进行反应——时间实验。最后一种实验的例子是：赫伯特·克拉克及其他人发现，当给受试者显示一个简单的图案时，比如加号上面的一颗星，然后在它旁边写上一个正确的陈述句（“星在加号上面。”）或者一个真实的否定句（“星不在加号下面。”）他们说第二个否定句是正确的时间，比说第一个肯定句是正确的时间长十分之二秒。我们好像更习惯于思考什么东西是什么，而不太习惯于思考什么东西不是什么，为了处理这些句子，我们得首先把否定句改写成肯定句。 今天，许多心理语言学家研究到最后都相信，环境的确给语言获取带来很大的影响，这与乔姆斯基所认为的要多些。比如，他们强调，由“母亲亲昵之语”，即一些母亲（包括一些父亲）对小孩子谈话的特别方式，进行的非正式语言培训。然而，尽管许多心理语言学家都对乔姆斯基的L． A． D．学说的一些细节提出了疑问（他本人对此也做了进一步的完善和修改），但他们中的大多数人都同意，人类具有基因决定的能力，以理解和获取任何语言。另外，按维克多利亚·弗罗姆肯的说法： 然而，这个问题仍然存在，即，这种基因决定的语言能力是否基本上是派生的－是否是总体认知的、生理学的和其它支持人类智力的非具体系统的副产品－或者，它是不是因为一些语言学上