Figure 1
Psychology is the science that investigates the representation and processing of information by complex organisms. Many animal species are capable of taking in information about their environment, forming internal representations of it, and manipulating these representations to select and execute actions. In addition, many animals are able to adapt to their environments by means of learning that can take place within the lifespan of an individual organism. Intelligent information processing implies the ability to acquire and process information about the environment in order to select actions that are likely to achieve the fundamental goals of survival and propagation. Animals have evolved a system of capabilities that collectively provide them with the ability to process information. They have sensory systems such as TASTE and HAPTIC PERCEPTION (touch), which provide information about the immediate environment with which the individual is in direct contact; proprioception, which provides information about an animal"s own bodily states; and SMELL, AUDITION, and VISION, which provide information about more distant aspects of the environment. Animals are capable of directed, self-generated motion, including EYE MOVEMENTS and other motoric behaviors such as MANIPULATION AND GRASPING, which radically increase their ability to pick up sensory information and also to act upon their environments.
The central focus of psychology concerns the information processing that intervenes between sensory inputs and motoric outputs. The most complex forms of intelligence, observed in birds and mammals, and particularly primates (especially great apes and humans) require theories that deal with the machinery of thought and inner experience. These animals have minds and EMOTIONS; their sensory inputs are interpreted to create perceptions of the external world, guided in part by selective ATTENTION; some of the products of perception are stored in MEMORY, and may in turn influence subsequent perception. Intellectually sophisticated animals perform DECISION MAKING and PROBLEM SOLVING, and in the case of humans engage in LANGUAGE AND COMMUNICATION. Experience coupled with innate constraints results in a process of COGNITIVE DEVELOPMENT as the infant becomes an adult, and also leads to LEARNING over the lifespan, so that the individual is able to adapt to its environment within a vastly shorter time scale than that required for evolutionary change. Humans are capable of the most complex and most domain-general forms of information processing of all species; for this reason (and because those who study psychology are humans), most of psychology aims directly or indirectly to understand the nature of human information processing and INTELLIGENCE. The most general characteristics of the human system for information processing are described as the COGNITIVE ARCHITECTURE.
As the science of the representation and processing of information by organisms, psychology (particularly cognitive psychology) forms part of the core of cognitive science. Cognitive science research conducted in other disciplines generally has actual or potential implications for psychology. Not all research on intelligent information processing is relevant to psychology. Some work in artificial intelligence, for example, is based on representations and algorithms with no apparent connection to biological intelligence. Even though such work may be highly successful at achieving high levels of competence on cognitive tasks, it does not fall within the scope of cognitive science. For example, the Deep Blue II program that defeated the human CHESS champion Gary Kasparov is an example of an outstanding artificial-intelligence program that has little or no apparent psychological relevance, and hence would not be considered to be part of cognitive science. In contrast, work on adaptive PRODUCTION SYSTEMS and NEURAL NETWORKS, much of which is conducted by computer scientists, often has implications for psychology. Similarly, a great deal of work in such allied disciplines as neuroscience, linguistics, anthropology, and philosophy has psychological implications. At the same time, work in psychology often has important implications for research in other disciplines. For example, research in PSYCHOLINGUISTICS has influenced developments in linguistics, and research in PSYCHOPHYSICS has guided neurophysiological research on the substrates of sensation and perception.
In terms of MARR's tripartite division of levels of analysis (computational theory, representation and algorithm, and hardware implementation), work in psychology tends to concentrate on the middle level, emphasizing how information is represented and processed by humans and other animals. Although there are many important exceptions, psychologists generally aim to develop process models that specify more than the input-output functions that govern cognition (for example, also specifying timing relations among intervening mental processes), while abstracting away from the detailed neural underpinnings of behavior. Nonetheless, most psychologists do not insist in any strict sense on the AUTONOMY OF PSYCHOLOGY, but rather focus on important interconnections with allied disciplines that comprise cognitive science. Contemporary psychology at the information-processing level is influenced by research in neuroscience that investigates the neural basis for cognition and emotion, by work on representations and algorithms in the fields of artificial intelligence and neural networks, and by work in social sciences such as anthropology that places the psychology of individuals within its cultural context. Research on the psychology of language (e.g., COMPUTATIONAL PSYCHOLINGUISTICS and LANGUAGE AND THOUGHT) is influenced by the formal analyses of language developed in linguistics. Many areas of psychology make close contact with classical issues in philosophy, especially in EPISTEMOLOGY (e.g., CAUSAL REASONING; INDUCTION; CONCEPTS).
The field of psychology has several major subdivisions, which have varying degrees of connection to cognitive science. Cognitive psychology deals directly with the representation and processing of information, with greatest emphasis on cognition in adult humans; the majority of the psychology entries that appear in this volume reflect work in this area. Developmental psychology deals with the changes in cognitive, social, and emotional functioning that occur over the lifespan of humans and other animals (see in particular COGNITIVE DEVELOPMENT, PERCEPTUAL DEVELOPMENT, and INFANT COGNITION). Social psychology investigates the cognitive and emotional factors involved in interactions between people, especially in small groups. One subarea of social psychology, SOCIAL COGNITION, is directly concerned with the manner in which people understand the minds, emotions, and behavior of themselves and others (see also THEORY OF MIND; INTERSUBJECTIVITY). Personality psychology deals primarily with motivational and emotional aspects of human experience (see FREUD for discussion of the ideas of the famous progenitor of this area of psychology), and clinical psychology deals with applied issues related to mental health. COMPARATIVE PSYCHOLOGY investigates the commonalities and differences in cognition and behavior between different animal species (see PRIMATE COGNITION; ANIMAL NAVIGATION; CONDITIONING; and MOTIVATION), and behavioral neuroscience provides the interface between research on molar cognition and behavior and their underlying neural substrate.
Until the middle of the nineteenth century the nature of the mind was solely the concern of philosophers. Indeed, there are a number of reasons why some have argued that the scientific investigation of the mind may prove to be an impossible undertaking. One objection is that thoughts cannot be measured; and without measurement, science cannot even begin. A second objection is to question how humans could objectively study their own thought processes, given the fact that science itself depends on human thinking. A final objection is that our mental life is incredibly complex and bound up with the further complexities of human social interactions; perhaps cognition is simply too complex to permit successful scientific investigation.
Despite these reasons for skepticism, scientific psychology emerged as a discipline separate from philosophy in the second half of the nineteenth century. A science depends on systematic empirical methods for collecting observations and on theories that interpret these observations. Beginning around 1850, a number of individuals, often trained in philosophy, physics, physiology, or neurology, began to provide these crucial elements.
The anatomist Ernst Heinrich Weber and the physicist and philosopher Gustav Fechner measured the relations between objective changes in physical stimuli, such as brightness or weight, and subjective changes in the internal sensations the stimuli generate. The crucial finding of Weber and Fechner was that subjective differences were not simply equivalent to objective differences. Rather, it turned out that for many dimensions, the magnitude of change required to make a subjective difference ("just noticeable difference," or "jnd") increased as overall intensity increased, often following an approximately logarithmic function, known as the Weber-Fechner Law. Weber and Fechner"s contribution to cognitive psychology was much more general than identifying the law that links their names. They convincingly demonstrated that, contrary to the claim that thought is inherently impossible to measure, it is in fact possible to measure mental concepts, such as the degree of sensation produced by a stimulus. Fechner called this new field of psychological measurement PSYCHOPHYSICS: the interface of psychology and physics, of the mental and the physical.
A further foundational issue concerns the speed of human thought. In the nineteenth century, many believed that thought was either instantaneous or else so fast that it could never be measured. But HERMANN VON HELMHOLTZ, a physicist and physiologist, succeeded in measuring the speed at which signals are conducted through the nervous system. He first experimented on frogs by applying an electric current to the top of a frog's leg and measuring the time it took the muscle at the end to twitch in response. Later he used a similar technique with humans, touching various parts of a person's body and measuring the time taken to press a button in response. The response time increased with the distance of the stimulus (i.e., the point of the touch) from the finger that pressed the button, in proportion to the length of the neural path over which the signal had to travel. Helmholtz's estimate of the speed of nerve signals was close to modern estimates -- roughly 100 meters per second for large nerve fibers. This transmission rate is surprisingly slow -- vastly slower than the speed of electricity through a wire. Because our brains are composed of neurons, our thoughts cannot be generated any faster than the speed at which neurons communicate with each other. It follows that the speed of thought is neither instantaneous nor immeasurable.
Helmholtz also pioneered the experimental study of vision, formulating a theory of color vision that remains highly influential today. He argued forcefully against the commonsensical idea that perception is simply a matter of somehow "copying" sensory input into the brain. Rather, he pointed out that even the most basic aspects of perception require major acts of construction by the nervous system. For example, it is possible for two different objects -- a large object far away, and a small object nearby -- to create precisely the same image on the retinas of a viewer's eyes. Yet normally the viewer will correctly perceive the one object as being larger, but further away, than the other. The brain somehow manages to unconsciously perform some basic geometrical calculations. The brain, Helmholtz argued, must construct this unified view by a process of "unconscious inference" -- a process akin to reasoning without awareness. Helmholtz's insight was that the "reality" we perceive is not simply a copy of the external world, but rather the product of the constructive activities of the brain.
Another philosopher, HERMANN EBBINGHAUS, who was influenced by Fechner's ideas about psychophysical measurements, developed experimental methods tailored to the study of human memory. Using himself as a subject, Ebbinghaus studied memory for nonsense syllables -- consonant-vowel-consonant combinations, such as "zad," "bim," and "sif." He measured how long it took to commit lists of nonsense syllables to memory, the effects of repetition on how well he could remember the syllables later, and the rate of forgetting as a function of the passage of time. Ebbinghaus made several fundamental discoveries about memory, including the typical form of the "forgetting curve" -- the gradual, negatively accelerated decline in the proportion of items that can be recalled as a function of time. Like Weber, Fechner, and Helmholtz, Ebbinghaus provided evidence that it is indeed possible to measure mental phenomena by objective experimental procedures.
Many key ideas about possible components of cognition were systematically presented by the American philosopher WILLIAM JAMES in the first great psychology textbook, Principles of Psychology, published in 1890. His monumental work included topics that remain central in psychology, including brain function, perception, attention, voluntary movement, habit, memory, reasoning, the SELF, and hypnosis. James discussed the nature of "will," or mental effort, which remains one of the basic aspects of attention. He also drew a distinction between different memory systems: primary memory, which roughly corresponds to the current contents of consciousness, and secondary memory, which comprises the vast store of knowledge of which we are not conscious at any single time, yet continually draw upon. Primary memory is closely related to what we now term active, short-term, or WORKING MEMORY, while secondary memory corresponds to what is usually called long-term memory.
James emphasized the adaptive nature of cognition: the fact that perception, memory, and reasoning operate not simply for their own sake, but to allow us to survive and prosper in our physical and social world. Humans evolved as organisms skilled in tool use and in social organization, and it is possible (albeit a matter of controversy) that much of our cognitive apparatus evolved to serve these basic functions (see EVOLUTIONARY PSYCHOLOGY). Thus, human cognition involves intricate systems for MOTOR CONTROL and MOTOR LEARNING; the capacity to understand that other people have minds, with intentions and goals that may lead them to help or hinder us; and the ability to recognize and remember individual persons and their characteristics. Furthermore, James (1890:8) recognized that the hallmark of an intelligent being is its ability to link ends with means -- to select actions that will achieve goals: "The pursuance of future ends and the choice of means for their attainment are thus the mark and criterion of the presence of mentality in a phenomenon." This view of goal-directed thinking continues to serve as the foundation of modern work on PROBLEM SOLVING, as reflected in the views of theorists such as ALAN NEWELL and Herbert Simon.
Another pioneer of psychology was Sigmund Freud, the founder of psychoanalysis, whose theoretical ideas about cognition and consciousness anticipated many key aspects of the modern conception of cognition. Freud attacked the idea that the "self" has some special status as a unitary entity that somehow governs our thought and action. Modern cognitive psychologists also reject (though for different reasons) explanations of intelligent behavior that depend upon postulating a "homunculus" -- that is, an internal mental entity endowed with all the intelligence we are trying to explain. Behavior is viewed not as the product of a unitary self or homunculus, but as the joint product of multiple interacting subsystems. Freud argued that the "ego" -- the information-processing system that modulates various motivational forces -- is not a unitary entity, but rather a complex system that includes attentional bottlenecks, multiple memory stores, and different ways of representing information (e.g., language, imagery, and physiognomic codes, or "body language"). Furthermore, as Freud also emphasized, much of information processing takes place at an unconscious level. We are aware of only a small portion of our overall mental life, a tip of the cognitive iceberg. For example, operating beneath the level of awareness are attentional "gates" that open or close to selectively attend to portions of the information that reaches our senses, memory stores that hold information for very brief periods of time, and inaccessible memories that we carry with us always but might never retrieve for years at a time.
Given the breadth and depth of the contributions of the nineteenth-century pioneers to what would eventually become cognitive science, it is ironic that early in the twentieth century the study of cognition went into a steep decline. Particularly in the United States, psychology in the first half of the century came to be dominated by BEHAVIORISM, an approach characterized by the rejection of theories that depended on "mentalistic" concepts such as goals, intentions, or plans. The decline of cognitive psychology was in part due to the fact that a great deal of psychological research had moved away from the objective measurement techniques developed by Fechner, Helmholtz, Ebbinghaus, and others, and instead gave primacy to the method of INTROSPECTION, promoted by WILHELM WUNDT, in which trained observers analyzed their own thought processes as they performed various cognitive tasks. Not surprisingly, given what is now known about how expectancies influence the way we think, introspectionists tended to find themselves thinking in more or less the manner to which they were theoretically predisposed. For example, researchers who believed thinking always depended on IMAGERY usually found themselves imaging, whereas those who did not subscribe to such a theory were far more likely to report "imageless thought."
The apparent subjectivity and inconstancy of the introspective method encouraged charges that all cognitive theories (rather than simply the method itself, as might seem more reasonable) were "unscientific." Cognitive theories were overshadowed by the behaviorist theories of such leading figures as John Watson, Edward Thorndike, Clark Hull, and B. F. Skinner. Although there were major differences among the behaviorists in the degree to which they actually avoided explanations based on assumptions about unobservable mental states (e.g., Hull postulated such states rather freely, whereas Watson was adamant that they were scientifically illicit), none supported the range of cognitive ideas advanced in the nineteenth century.
Cognitive psychology did not simply die out during the era of behaviorism. Working within the behaviorist tradition, Edward Tolman pursued such cognitive issues as how animals represented spatial information internally as COGNITIVE MAPS of their environment. European psychologists were far less captivated with behaviorism than were Americans. In England, Sir FREDERICK BARTLETT analyzed the systematic distortions that people exhibit when trying to remember stories about unfamiliar events, and introduced the concept of "schema" (see SCHEMATA) as a mental representation that captures the systematic structural relations in categories of experience. In Soviet Russia, the neuropsychologist Aleksandr LURIA provided a detailed portrait of links between cognitive functions and the operation of specific regions of the brain. Another Russian, LEV VYGOTSKY, developed a sociohistorical approach to cognitive development that emphasized the way in which development is constructed through social interaction, cultural practices, and the internalization of cognitive tools. Vygotsky emphasized social interaction through language in the development of children's concepts. The Swiss psychologist JEAN PIAGET spent decades refining a theory of cognitive development. Piaget"s theory emphasizes milestones in the child's development including decentration, the ability to perform operations on concrete objects, and finally the ability to perform operations on thoughts and beliefs. Given its emphasis on logical thought, Piaget"s theory is closely related to SCIENTIFIC THINKING AND ITS DEVELOPMENT .
In addition, the great German tradition in psychology, which had produced so many of the nineteenth-century pioneers, gave rise to a new cognitive movement in the early twentieth century: GESTALT PSYCHOLOGY. The German word Gestalt translates roughly as "form," and the Gestalt psychologists emphasized that the whole form is something different from the mere sum of its parts, due to emergent properties that arise as new relations are created. Gestalt psychology was in some ways an extension of Helmholtz"s constructivist ideas, and the greatest contributions of this intellectual movement were in the area of GESTALT PERCEPTION. Where the behaviorists insisted that psychology was simply the study of how objective stimuli come to elicit objective responses, the Gestaltists pointed to simple demonstrations casting doubt on the idea that "objective" stimuli -- that is, stimuli perceived in a way that can be described strictly in terms of the sensory input -- even exist. Figure 1 illustrates a famous Gestalt example of the constructive nature of perception, the ambiguous Necker cube. Although this figure is simply a flat line drawing, we immediately perceive it as a three-dimensional cube. Moreover, if you look carefully, you will see that the figure can actually be seen as either of two different three-dimensional cubes. The same objective stimulus -- the two-dimensional line drawing -- gives rise to two distinct three-dimensional perceptions.
Figure 1
Although many of the major contributions by key Gestalt figures such as Max Wertheimer were in the area of perception, their central ideas were extended to memory and problem solving as well, through the work of people such as Wolfgang Köhler and Karl Duncker. Indeed, one of the central tenets of Gestalt psychology was that high-level thinking is based on principles similar to those that govern basic perception. As we do in everyday language, Gestalt psychologists spoke of suddenly "seeing" the solution to a problem, often after "looking at it" in a different way and achieving a new "insight." In all the areas in which they worked, the Gestalt idea of "a whole different from the sum of parts" was based on the fundamental fact that organized configurations are based not simply on individual elements, but also on the relations between those elements. Just as H2O is not simply two hydrogen atoms and one oxygen atom, but also a particular spatial organization of these elements into a configuration that makes a molecule of water, so too "squareness" is more than four lines: it crucially depends on the way the lines are related to one another to make four right angles. Furthermore, relations can take on a "life of their own," separable from any particular set of elements. For example, we can take a tune, move it to a different key so that all the notes are changed, and still immediately recognize it as the "same" tune as long as the relations among the notes are preserved. A focus on relations calls attention to the centrality of the BINDING PROBLEM, which involves the issue of how elements are systematically organized to fill relational roles. Modern work on such topics as ANALOGY and SIMILARITY emphasizes the crucial role of relations in cognition.
Modern cognitive psychology emerged in the second half of this century. The "cognitive revolution" of the 1950s and 1960s involved not only psychology but also the allied disciplines that now contribute to cognitive science. In the 1940s the Canadian psychologist DONALD HEBB began to draw connections between cognitive processes and neural mechanisms, anticipating modern cognitive neuroscience. During World War II, many experimental psychologists (including JAMES GIBSON) were confronted with such pressing military problems as finding ways to select good pilots and train radar operators, and it turned out that the then-dominant stimulus-response theories simply had little to offer in the way of solutions. More detailed process models of human information processing were needed. After the war, DONALD BROADBENT in England developed the first such detailed model of attention. Even more importantly, Broadbent helped develop and popularize a wide range of experimental tasks in which an observer"s attention is carefully controlled by having him or her perform some task, such as listening to a taped message for a particular word, and then precisely measuring how quickly responses can be made and what can be remembered. In the United States, William K. Estes added to the mathematical tools available for theory building and data analysis, and Saul Sternberg developed a method for decomposing reaction times into component processes using a simple recognition task.
Meanwhile, the birth of computer science provided further conceptual tools. Strict behaviorists had denounced models of internal mental processes as unscientific. However, the modern digital computer provided a clear example of a device that took inputs, fed them through a complex series of internal procedures, and then produced outputs. As well as providing concrete examples of what an information-processing device could be, computers made possible the beginnings of artificial intelligence -- the construction of computer programs designed to perform tasks that require intelligence, such as playing chess, understanding stories, or diagnosing diseases. Herbert Simon (1978 Nobel Laureate in Economics) and Allan Newell were leaders in building close ties between artificial intelligence and the new cognitive psychology. It was also recognized that actual computers represent only a small class of a much larger set of theoretically possible computing devices, which had been described back in the 1940s by the brilliant mathematician ALAN TURING. Indeed, it was now possible to view the brain itself as a biological computer, and to use various real and possible computing devices as models of human cognition. Another key influence on modern cognitive psychology came from the field of linguistics. In the late 1950s work by the young linguist Noam Chomsky radically changed conceptions of the nature of human language by demonstrating that language could not be learned or understood by merely associating adjacent words, but rather required computations on abstract structures that existed in the minds of the speaker and listener.
The collective impact of this work in the mid-twentieth century was to provide a seminal idea that became the foundation of cognitive psychology and also cognitive science in general: the COMPUTATIONAL THEORY OF MIND, according to which human cognition is based on mental procedures that operate on abstract mental representations. The nature of the COGNITIVE ARCHITECTURE has been controversial, including proposals such as PRODUCTION SYSTEMS and NEURAL NETWORKS. In particular, there has been disagreement as to whether procedures and representations are inherently separable or whether procedures actually embody representations, and whether some mental representations are abstract and amodal, rather than tied to specific perceptual systems. Nonetheless, the basic conception of biological information processing as some form of computation continues to guide psychological theories of the representation and processing of information.
In broad strokes, an intelligent organism operates in a perception-action cycle (Neisser 1967), taking in sensory information from the environment, performing internal computations on it, and using the results of the computation to guide the selection and execution of goal-directed actions. The initial sensory input is provided by separate sensory systems, including smell, taste, haptic perception, and audition. The most sophisticated sensory system in primates is vision (see MID-LEVEL VISION; HIGH-LEVEL VISION), which includes complex specialized subsystems for DEPTH PERCEPTION, SHAPE PERCEPTION, LIGHTNESS PERCEPTION, and COLOR VISION.
The interpretation of sensory inputs begins with FEATURE DETECTORS that respond selectively to relatively elementary aspects of the stimulus (e.g., lines at specific orientations in the visual field, or phonetic cues in an acoustic speech signal). Some basic properties of the visual system result in systematic misperceptions, or ILLUSIONS. TOP-DOWN PROCESSING IN VISION serves to integrate the local visual input with the broader context in which it occurs, including prior knowledge stored in memory. Theorists working in the tradition of Gibson emphasize that a great deal of visual information may be provided by higher-order features that become available to a perceiver moving freely in a natural environment, rather than passively viewing a static image (see ECOLOGICAL PSYCHOLOGY). In their natural context, both perception and action are guided by the AFFORDANCES of the environment: properties of objects that enable certain uses (e.g., the elongated shape of a stick may afford striking an object otherwise out of reach).
Across all the sensory systems, psychophysics methods are used to investigate the quantitative functions relating physical inputs received by sensory systems to subjective experience (e.g., the relation between luminance and perceived brightness, or between physical and subjective weight). SIGNAL DETECTION THEORY provides a statistical method for measuring how accurately observers can distinguish a signal from noise under conditions of uncertainty (i.e., with limited viewing time or highly similar alternatives) in a way that separates the signal strength received from possible response bias. In addition to perceiving sensory information about objects at locations in space, animals perceive and record information about time (see TIME IN THE MIND).
Knowledge about both space and time must be integrated to provide the capability for animal and HUMAN NAVIGATION in the environment. Humans and other animals are capable of forming sophisticated representations of spatial relations integrated as COGNITIVE MAPS. Some more central mental representations appear to be closely tied to perceptual systems. Humans use various forms of imagery based on visual, auditory and other perceptual systems to perform internal mental processes such as MENTAL ROTATION. The close connection between PICTORIAL ART AND VISION also reflects the links between perceptual systems and more abstract cognition.
A fundamental property of biological information processing is that it is capacity-limited and therefore necessarily selective. Beginning with the seminal work of Broadbent, a great deal of work in cognitive psychology has focused on the role of attention in guiding information processing. Attention operates selectively to determine what information is received by the senses, as in the case of EYE MOVEMENTS AND VISUAL ATTENTION, and also operates to direct more central information processing, including the operation of memory. The degree to which information requires active attention or memory resources varies, decreasing with the AUTOMATICITY of the required processing.
Modern conceptions of memory maintain some version of William James's basic distinction between primary and secondary memory. Primary memory is now usually called WORKING MEMORY, which is itself subdivided into multiple stores involving specific forms of representation, especially phonological and visuospatial codes. Working memory also includes a central executive, which provides attentional resources for strategic management of the cognitive processes involved in problem solving and other varieties of deliberative thought. Secondary or long-term memory is also viewed as involving distinct subsystems, particularly EPISODIC VS. SEMANTIC MEMORY. Each of these subsystems appears to be specialized to perform one of the two basic functions of long-term memory. One function is to store individuated representations of "what happened when" in specific contexts (episodic memory); a second function is to extract and store generalized representations of "the usual kind of thing" (semantic memory). Another key distinction, related to different types of memory measures, is between IMPLICIT VS. EXPLICIT MEMORY. In explicit tests (typically recall or recognition tests), the person is aware of the requirement to access memory. In contrast, implicit tests (such as completing a word stem, or generating instances of a category) make no reference to any particular memory episode. Nonetheless, the influence of prior experiences may be revealed by the priming of particular responses (e.g., if the word "crocus" has recently been studied, the person is more likely to generate "crocus" when asked to list flowers, even if they do not explicitly remember having studied the word). There is evidence that implicit and explicit knowledge are based on separable neural systems. In particular, forms of amnesia caused by damage to the hippocampus and related structures typically impair explicit memory for episodes, but not implicit memory as revealed by priming measures.
A striking part of human cognition is the ability to speak and comprehend language. The psychological study of language, or psycholinguistics, has a close relationship to work in linguistics and on LANGUAGE ACQUISITION. The complex formal properties of language, together with its apparent ease of acquisition by very young children, have made it the focus of debates about the extent and nature of NATIVISM in cognition. COMPUTATIONAL PSYCHOLINGUISTICS is concerned with modeling the complex processes involved in language use. In modern cultures that have achieved LITERACY with the introduction of written forms of language, the process of READING lies at the interface of psycholinguistics, perception, and memory retrieval. The intimate relationship between language and thought, and between language and human concepts, is widely recognized but still poorly understood. The use of METAPHOR in language is related to other symbolic processes in human cognition, particularly ANALOGY and CATEGORIZATION.
One of the most fundamental aspects of biological intelligence is the capacity to adaptively alter behavior. It has been clear at least from the time of William James that the adaptiveness of human behavior and the ability to achieve EXPERTISE in diverse domains is not generally the direct product of innate predispositions, but rather the result of adaptive problem solving and LEARNING SYSTEMS that operate over the lifespan. Both production systems and neural networks provide computational models of some aspects of learning, although no model has captured anything like the full range of human learning capacities. Humans as well as some other animals are able to learn by IMITATION, for example, translating visual information about the behavior of others into motor routines that allow the observer/imitator to produce comparable behavior. Many animal species are able to acquire expectancies about the environment and the consequences of the individual"s actions on the basis of CONDITIONING, which enables learning of contingencies among events and actions.
Conditioning appears to be a primitive form of causal induction, the process by which humans and other animals learn about the cause-effect structure of the world. Both causal knowledge and similarity relations contribute to the process of categorization, which leads to the development of categories and concepts that serve to organize knowledge. People act as if they assume the external appearances of category members are caused by hidden (and often unknown) internal properties (e.g., the appearance of an individual dog may be attributed to its internal biology), an assumption sometimes termed psychological ESSENTIALISM.
There are important developmental influences that lead to CONCEPTUAL CHANGE over childhood. These developmental aspects of cognition are particularly important in understanding SCIENTIFIC THINKING AND ITS DEVELOPMENT. Without formal schooling, children and adults arrive at systematic beliefs that comprise NAIVE MATHEMATICS and NAIVE PHYSICS. Some of these beliefs provide the foundations for learning mathematics and physics in formal EDUCATION, but some are misconceptions that can impede learning these topics in school (see also AI AND EDUCATION). Young children are prone to ANIMISM, attributing properties of people and other animals to plants and nonliving things. Rather than being an aberrant form of early thought, animism may be an early manifestation of the use of ANALOGY to make inferences and learn new cognitive structures. Analogy is the process used to find systematic structural correspondences between a familiar, well-understood situation and an unfamiliar, poorly understood one, and then using the correspondences to draw plausible inferences about the less familiar case. Analogy, along with hypothesis testing and evaluation of competing explanations, plays a role in the discovery of new regularities and theories in science.
In its more complex forms, learning is intimately connected to thinking and reasoning. Humans are not only able to think, but also to think about their own cognitive processes, resulting in METACOGNITION. They can also form higher-level representations, termed METAREPRESENTATION. There are major individual differences in intelligence as assessed by tasks that require abstract thinking. Similarly, people differ in their CREATIVITY in finding solutions to problems. Various neural disorders, such as forms of MENTAL RETARDATION and AUTISM, can impair or radically alter normal thinking abilities. Some aspects of thinking are vulnerable to disruption in later life due to the links between AGING AND COGNITION.
Until the last few decades, the psychology of DEDUCTIVE REASONING was dominated by the view that human thinking is governed by formal rules akin to those used in LOGIC. Although some theorists continue to argue for a role for formal, content-free rules in reasoning, others have focused on the importance of content-specific rules. For example, people appear to have specialized procedures for reasoning about broad classes of pragmatically important tasks, such as understanding social relations or causal relations among events. Such pragmatic reasoning schemas (Cheng and Holyoak 1985) enable people to derive useful inferences in contexts related to important types of recurring goals. In addition, both deductive and inductive inferences may sometimes be made using various types of MENTAL MODELS, in which specific possible cases are represented and manipulated (see also CASE-BASED REASONING AND ANALOGY).
Much of human inference depends not on deduction, but on inductive PROBABILISTIC REASONING under conditions of UNCERTAINTY. Work by researchers such as AMOS TVERSKY and Daniel Kahneman has shown that everyday inductive reasoning and decision making is often based on simple JUDGMENT HEURISTICS related to ease of memory retrieval (the availability heuristic) and degree of similarity (the representativeness heuristic). Although judgment heuristics are often able to produce fast and accurate responses, they can sometimes lead to errors of prediction (e.g., conflating the subjective ease of remembering instances of a class of events with their objective frequency in the world).
More generally, the impressive power of human information processing has apparent limits. People all too often take actions that will not achieve their intended ends, and pursue short-term goals that defeat their own long-term interests. Some of these mistakes arise from motivational biases, and others from computational limitations that constrain human attention, memory, and reasoning processes. Although human cognition is fundamentally adaptive, we have no reason to suppose that "all's for the best in this best of all possible minds."
Cheng, P. W., and K. J. Holyoak. (1985). Pragmatic reasoning schemas. Cognitive Psychology 17:391-394.
James, W. (1890). The Principles of Psychology. New York: Dover.
Neisser, U. (1967). Cognitive Psychology. Englewood Cliffs, NJ: Prentice-Hall.
Anderson, J. R. (1995). Cognitive Psychology and Its Implications. 4th ed. San Francisco: W. H. Freeman.
Baddeley, A. D. (1997). Human Memory: Theory and Practice. 2nd ed. Hove, Sussex: Psychology Press.
Evans, J., S. E. Newstead, and R. M. J. Byrne. (1993). Human Reasoning. Mahwah, NJ: Erlbaum.
Gallistel, C. R. (1990). The Organization of Learning. Cambridge, MA: MIT Press.
Gazzaniga, M. S. (1995). The Cognitive Neurosciences. Cambridge, MA: MIT Press.
Gibson, J. J. (1979). The Ecological Approach to Visual Perception. Boston: Houghton-Mifflin.
Gregory, R. L. (1997). Eye and Brain:The Psychology of Seeing. 5th ed. Princeton, NJ: Princeton University Press.
Holyoak, K. J., and P. Thagard. (1995). Mental Leaps: Analogy in Creative Thought. Cambridge, MA: MIT Press.
James, W. (1890). Principles of Psychology. New York: Dover.
Kahneman, D., P. Slovic, and A. Tversky. (1982). Judgments Under Uncertainty: Heuristics and Biases. New York: Cambridge University Press.
Keil, F. C. (1989). Concepts, Kinds, and Cognitive Development. Cambridge, MA: MIT Press.
Kosslyn, S. M. (1994). Image and Brain: The Resolution of the Imagery Debate. Cambridge, MA: MIT Press.
Newell, A., and H. A. Simon. (1972.) Human Problem Solving. Englewood Cliffs, NJ: Prentice-Hall.
Pashler, H. (1997). The Psychology of Attention. Cambridge, MA: MIT Press.
Pinker, S. (1994). The Language Instinct. New York: William Morrow.
Reisberg, D. (1997). Cognition: Exploring the Science of the Mind. New York: Norton.
Rumelhart, D. E., J. L. McClelland, and PDP Research Group. (1986). Parallel Distributed Processing: Explorations in the Microstructure of Cognition. 2 vols. Cambridge, MA: MIT Press.
Smith, E. E., and D. L. Medin. (1981). Categories and Concepts. Cambridge, MA: Harvard University Press.
Sperber, D., D. Premack, and A. J. Premack. (1995). Causal Cognition: A Multidisciplinary Debate. Oxford: Clarendon Press.
Tarpy, R. M. (1997). Contemporary Learning Theory and Research. New York: McGraw Hill.
Tomasello, M., and J. Call. (1997). Primate Cognition. New York: Oxford University Press.
Copyright © 1999 Massachusetts Institute of Technology