Intellectual development refers here to the changes that occur, as a result of growth and experience, in a person’s capacities for thinking, reasoning, relating, judging, conceptualizing, etc. In particular it concerns such changes in children.
There are a number of different approaches to the study of intellectual development in children. As in the history of most branches of scientific knowledge, the study began with observation and description. For many years descriptive accounts of children’s thinking, reasoning, and other intellectual capacities were thoroughly mixed with descriptions of their social and emotional development and of their verbal and motor skills. Moreover, there was at first a tendency to attribute to the child mental processes that were simply miniature versions of adult thought patterns. Such early observers as Darwin (1877) were careful and deliberate, but their records often revealed the limitations of studying only one child, and the biases of the observer.
Predictably, the early, unsystematic observation of one child at a time was eventually replaced by systematic efforts to measure children’s behavior and capacities in standardized and objective ways. The growth of the mental testing movement in the first 40 years of the twentieth century testifies to the enthusiasm that was generated by the possibility of applying the precision of quantitative measurement to the task of comparing individual children and calibrating the changes that take place over the early years of life. Although observation had been supplemented by measurement, the primary purpose of these efforts remained descriptive, and the generalizations achieved were themselves only descriptions of trends and improvements that occurred consistently with increasing age.
Still more recently, since about 1950, there has been an increasing movement toward the laboratory study of the ways in which patterns of development themselves change as age changes. This recent work has been not so much concerned with the effects of age itself as with the development in children of certain functional relationships between experience and performance that have been demonstrated in human adults and have been found lacking in most infrahuman species. The emphasis is on the application of laboratory controls and experimental manipulations to the study of cognitive development. The aim is to control the stimulus conditions under which behavior is observed and to explain why intellect develops, as well as describing how and when it develops.
Such an approach does not obviate the need for study of the child’s understanding as it changes with age. Rather, it relies on developmental descriptions of intellectual processes and products for clues as to when a certain level of understanding or specific intellectual accomplishment is likely to be achieved, and what repertoire of cognitive processes constitutes the means available for such an accomplishment at that age. Even the correlation of processes with products over ages, however, leaves the detailed cause-effect analysis still to be performed.
Although the present article is not primarily concerned with age changes per se, it should be noted that the description of age changes in intellectual functioning continues to thrive in two lines of research. One is the continued development and refinement of standardized tests of intelligence in order to predict an individual’s future intellectual achievement and to select, train, and guide children whenever a test-derived forecast can aid in making decisions on their behalf [seeIntelligence AND Intelligence Testing].
The second line of research is that of Jean Piaget and his associates on cognitive development. This large body of work has been concerned with the ontogenetic unfolding and evolution of cognitive capacities in the child, and like the work of Heinz Werner (1926), it has an organic quality and a complexity that are quite different from the empirical, item-analysis tradition of the test developers.
Both of these lines of research are structural in emphasis, i.e., they are primarily concerned with identifying the component parts or capacities of the intellect and with the organization and arrangement of these parts. The test developers are concerned with objective measurement of capacities in quantitative terms governed by a sophisticated statistics and a well-worked-out theory of measurement. The genetic epistemologists, on the other hand, have followed Piaget’s lead in attempting to describe the step-by-step development of the child’s understanding of his world as it progresses toward a formal, abstract, and logical comprehension of operations and relations in that world. Recent investigators stimulated by Piaget’s work have begun the task of isolating the conditions necessary for cognitive change and the explication of processes as well as products. [SeeDevelopmental Psychology, especially the article on A Theory OF Development.]
In contrast, a functional emphasis, i.e., a concern with dynamics, processes, and interrelationships, is found in the descriptions of cognitive development and in the explorations of dynamic mechanisms in cognitive change that have largely been undertaken by American behaviorists and behavior analysts and by Soviet pedagogists. These lines of research are more concerned with the processes of learning and thinking than with the structure of understanding. It is to the contributions of these functionalists that the present article is primarily devoted. It will be necessary first to summarize the most important age changes that have been described from infancy to adolescence. Consideration is then given to cognition, seen as the elaboration and selective generalization of simpler forms of learning and conditioning. Concepts such as mediation, learning set, and expectancy are discussed in relation to experimental studies of discrimination learning and discrimination reversal, concept formation, and the perceptual constancies. Curiosity and exploratory motivation are treated in relation to orienting responses and observing behavior. Research on acquired dis-tinctiveness, equivalence, and relevance of cues is presented as evidence for the importance of a general class of intervening responses, and the major role of language in this connection is stressed. Finally, consideration is given to individual differences in cognitive style, including discussion of such variables as field dependency, rigidity, reflectivity, and creativity.
Very little behavior in infancy possesses that degree of orderliness and abstractness which would qualify it as intellectual or cognitive. There are evident, however, the beginnings of systematic relations with the environment that imply understanding on a primitive level. Beyond the specific and identifiable reflexes, newborn behavior is usually described as massive, diffuse, and all-or-none in its occurrence. Recent investigations of neonatal behavior have shown that such descriptions mask at least seven reliably discriminable states of arousal, ranging from deep, unresponsive sleep to intense crying or sucking that is equally unresponsive to external stimulation. Reactivity to external stimuli is greatest at intermediate values on the arousal dimension. [SeeDrives, article onPhysiological Drives.]
Beyond the neonatal period (birth to ten days) there are three kinds of behavior in infants that appear to mark the beginnings of cognitive development: the development of simple stimulus equivalences, expectations, and persistent exploratory behavior.
Stimulus equivalence. Stimulus equivalence means perceptual recognition of the same object, person, or event under variable appearances due to changes in distance, illumination, context, angle of regard, and the like. While there have been no formal psychophysical investigations of perceptual constancies in infants, studies of attention and recognition of familiar people and objects have indicated that considerable equivalence learning takes place in early infancy (Rheingold et al. 1959). Infants also show selective attention, indicating some kind of differential sensitivity to faces and patterned stimuli (R. L. Fantz, quoted in Gibson 1963). [SeePerception, article on Perceptual Constancy.]
Expectations. If certain events occur in a brief, invariant sequence with considerable frequency and regularity in the infant’s environment, he is capable of displaying anticipatory responses that constitute for many psychologists sufficient evidence of expectancy. For example, it is commonly observed that quieting and sucking responses at first occur primarily in response to tactile stimulation of the mouth or cheek and general handling. Very soon, however, they begin to occur in response to vocalization by the mother and to other sights and sounds accompanying her approach. Another example is the eye blink, which at first occurs in response to tactile stimulation only and later occurs whenever a figure grows rapidly in size (“looms”), as if it were approaching the face at high speed. The acquisition of such simple preparatory responses as these does not require a cognitive interpretation and is explained satisfactorily by either classical or operant conditioning theories. But motoric anticipation does seem to provide a promising basis from which the child may develop more abstract capacities to anticipate the outcomes of his actions and to interpret signals of events.
Exploratory behavior. One of the most persistent and ubiquitous observations of infants of many species is their exploratory behavior, from which curiosity, both as a cognitive and as a motivational variable, is inferred. Persistent handling, mouthing, smelling, and visual scanning seem to occur in the first year or nine months in proportion to the familiarity of the stimuli, and in the first half of the second year in proportion to the relative novelty of the stimuli. It has been suggested that this apparent reversal merely reflects an increase with experience in the degree of variation required to make a stimulus relatively novel. In any case, the behavior is persistent and repetitive, and only toward the end of infancy is there evidence for selective matching of the child’s exploratory behavior to the nature of the object being explored. Before 18 months the repertoire of exploratory movements is small and undifferentiated. [SeeStimulation Drives.]
Intellectual development in the second and third years is dominated by the acquisition of communicative and referential language. Although it is not until age five or six that language becomes available as a conceptual system for the mediation of thinking, reasoning, and problem solving, the child of 1 ½ to 2 years begins to respond to verbal stimuli and to make verbal responses that are selectively appropriate to his immediate environment.
Communicative language. Acquisition of communicative language means the child’s learning to respond to simple instructions and demands made by others and his growing ability to verbalize his own need states and wishes with sufficient accuracy and clarity that others can respond appropriately. Most of this learning is acquired as part of training in daily routines, such as feeding, dressing, sleeping, and elimination. Not only are verbal demands made that require the child to discriminate the appropriate occasions and circumstances for each of these activities, but also he is expected to report with some accuracy on his internal need states, such as hunger, fatigue, pain, and need to eliminate. Because of the deliberate simplicity of the verbal demands made on him, the consistency with which they are made, and the immediacy of feedback from the child’s compliance (or noncom-pliance), this learning proceeds rapidly but remains at a concrete level. Similarly, the primacy and strength of the needs that the child is asked to verbalize and the usual promptness of their satisfaction ensure the accuracy, but limit the subtlety, of his verbal self-report. Typically, all that is required of him is a “Yes” or “No” answer or a choice of simple responses.
Referential language. Referential language consists of a collection of names for objects and events that enables the establishment of a simple representational system. The child of three can name common objects and actions and can pick objects out of an array by name. The development of this capacity is said to proceed from control of behavior by the gross motoric aspects of speech toward control by the semantic aspects of speech. This accomplishment in the third year depends, according to Luriia (1961), on establishing verbal control of the child’s attentional and orienting behavior. [Seeattention; language, article onLanguage Development; Learning, article OnVerbal Learning.]
Unverbalized comprehension. Progress in un-verbalized comprehension during the second and third years is less noticeable than is linguistic development. The child can interpret the social significance of an increased range of gestures and facial expressions, which by association come to signal more primary rewards and threats. He engages in considerable testing of the social consequences of his own behavior, reading these secondary cues in place of verbal feedback. Thus he is able to explore the socially imposed limitations of his own autonomy, interpreting the behavior of others so as to stop short of provoking severe thfeat or punishment. [SeePerception, article onSocial Perception.]
In the laboratory the child of less than three years cannot readily solve even simple discrimination and concept formation problems. His attention is diffusely focused and highly distractable. His representational system is concrete, egocentric, lacking in temporal perspective, and generally inflexible. His learning seems to proceed by simple associations, and despite his capacity for speech, the child rarely uses language to organize information or to direct his own behavior.
The child from three to five years makes intellectual progress primarily in establishing a frame of reference for his own identity within the family, internalizing a set of societal standards or behavior, copying a set of assumptions and beliefs about good and bad conduct, and assuming prescribed age and sex roles. The basic process of family socialization that began formally in the first year is substantially completed by the beginning of the sixth year in most Western cultures. During the preschool years the child gains considerable understanding of himself as one person in a world of people. He begins to be able to take the viewpoint of others and perceive the reciprocity of interpersonal encounters. [SeeIdentity, Psychosocial; Socialization.]
Objectivity and planning. While the process of socialization discussed above is not a pure instance of cognitive development, it furnishes evidence for, and practice with, a new relativity and objectivity in the child’s thought. His representational system is well enough developed so that he can deal with hypothetical cause and effect in situations that he has never experienced concretely. His play and fantasy give evidence of a high degree of realism, internal logic, and continuity. In simple situations he can reason backward from a desired state of affairs to the existing circumstances, and thus can increase the likelihood that his first actions will move him in the desired direction by simple but intelligent planning.
Discrimination learning. Dimensions of meaning other than the evaluative become available to the child. He can discriminate actions and events on such dimensions as fair-unfair, masculine-feminine, certain-uncertain, near-far in time, active-passive, and fast-slow. Simultaneous discrimination learning and simple concept learning are within his capacities, but successive discriminations and multivariate concepts remain difficult or impossible for him to learn. He can readily learn to make comparisons among stimuli on the basis of size, shape, color, texture, weight, material, speed, warmth, hardness, or numerosity, and in simple situations he can transfer these discriminations to new stimuli with some success. [SeeLearning, article onDiscrimination Learning.]
Language and verbal behavior. Language gains are extensive in the period from three to six years, but the progress appears more in pronunciation, grammar, and vocabulary than as an aid to abstraction and generalization. Language structure gains some independence of the words used, so that the child can shape sentences containing proper negation, tense, voice, person, number, and sometimes mood. For the first time, word order and invariant prefixes and suffixes seem to be acquired more as general rules than as rigid fragments of rote-learned verbal associations.
In spite of these demonstrated relationships between language and learning, there is good evidence that semantic verbal mediation as an aid to problem solving is neither typically nor fully available at this age. Studies indicate that while the child of five may sometimes solve the problem, his performance lacks that degree of reliance upon verbal mediation which characterizes the performance of a seven- or eight-year-old. Presumably for this reason, the younger child’s learning is not as rapid, flexible, consistent, and free of irrelevant responses as is that of the older, mediating child (see Kendler 1963). One might make such comparisons at any age, however, and it is likely that in general a curvilinear relationship will usually be found between reliance on verbal mediators and task difficulty, with the most mediation occurring on problems of intermediate difficulty.
Middle childhood. Between the ages of six and eleven the child’s capacity for thought and reasoning shows its most significant growth. To an undetermined degree this growth is stimulated by the beginning of formal scholastic instruction and the acquisition of reading and writing skills. But educators and psychologists generally agree that the maturation of intellectual capacities at about age six provides a readiness for formal learning and instruction. It has been claimed that there is a discontinuity in cognitive functioning at about age six, before which learning is basically a process of simple association and primary generalization. Before six, thinking is said to be a matter of recognition, literal recall, and direct transfer of training. After six, learning is said to be more focused and systematic, more abstract and selective, and much more flexibly applicable to a variety of contexts and contents. Thinking is more rapid, orderly, and precise. It is characterized by more insightful induction of generalities and more parsimonious application of general principles to a much greater range of content (White 1965).
Verbal mediation. Recent research indicates that the discontinuity in cognitive functioning is not so much an abrupt change at a specific time as it is a reflection of an underlying dichotomy between verbal and nonverbal approaches to problem solving. Thus, there may be a gradual shift over several years from reliance on nonverbal solutions to reliance on verbally mediated ones. Again, task difficulty relative to his present level will often determine whether a child in transition will attempt a verbal approach.
After this transition, language is employed to mediate all kinds of learning and problem solving, whether the child has a set to employ verbal mediation or not. Symbolic notation and graphic analogues become useful tools for problem solving and supplement the process of verbal reasoning. The child’s knowledge about the world becomes coded, hierarchically organized, and interconnected. Intact chains and clusters of associations or meanings can be applied to new problems at different levels of generality, in sharp contrast with the chaotic array of fragmentary facts and associations that characterized the knowledge of the younger child. Moreover, by age 11 the child learns to discriminate levels of abstraction and generality, relevance and importance of new information, regularities and patterns in sequential events, and intermediate degrees of likelihood between certainty and indeterminacy.
Role-taking and communication. There is a corresponding growth during middle childhood in role-taking and communications skills. The growing abstractness and relativity of the child’s understanding is applied to many new social situations, producing a complementary growth in his ability to understand and take into account the viewpoints of others in the course of his social interaction with them.
He understands concepts like fairness and sharing, although he applies them rigidly and indiscriminately. His perspective is broader and his experiences, now that they are codable, are consequently more readily available for use in making decisions. Not only is the child’s knowledge coded and ordered by means of verbal labels so that it can be selectively retrieved from memory, but also by middle childhood the child can use language to guide and monitor his own behavior through complex and contingently branching alternate plans from the definition of a problem through its solution.
Beyond age 10 to 11, development is more quantitative than qualitative. Existing capacities are refined and extended to new material. In particular the child’s ability to view problems in a variety of social contexts, to comprehend the variable sources of other people’s behavior, expands rapidly during adolescence. The capacity to see the world as another sees it and to act on the basis of a more sophisticated interpersonal relativity reaches full development only in late adolescence and young adulthood. There is, however, little evidence that the major personal reorientation that takes place during adolescence in many cultures has any necessary structural implications for cognitive development.
Research on children’s perception concerns itself mostly with psychophysical determination of absolute and differential thresholds, susceptibility to illusions and aftereffects, perceptual constancies, scaling of preferences, and capacities of blind and deaf children. (For a recent review of new research in perceptual development, see Pick et al. 1966.)
Constancies and concepts. There is a considerable body of literature on the development of the perceptual constancies in children (see Wohlwill 1960) and the field determination of perceptual experience as a function of age (Piaget et al. 1958). Because experience and knowledge are highly relevant to illusion susceptibility and the perceptual constancies, it is unfortunate that these relationships have not been more thoroughly explored, especially in the literature available in English. The perceptual constancies, for example, may be regarded as prototypic instances of concept formation. To the extent that an object is recognizably itself, regardless of distance, angle of regard, illumination, and other variable conditions of viewing, its successive appearances may be formally analyzed as different instances of a single concept. [SeePerception, articles on Perceptual Development, Perceptual CONSTANCY, and ILLUSIONS AND AFTEREFFECTS.]
The capacity to understand spoken language is another perceptual bridge to conceptual thought. Just as the child without clear intent or formal training comes to recognize the same object consistently through all its different appearances, so he comes to comprehend spoken language despite the many distortions and irrelevant variations that occur with different speakers and on different occasions.
For example, spectrographic analysis of speech indicates that certain critical features of phonemes must remain constant in order for an utterance to be accurately perceived, while certain aspects of pitch, vowel quality, and timing can vary widely without diminishing intelligibility. We infer that the child learns to select a small number of relevant cues from the complex of speech sounds that he hears. The major unsolved problem in speech perception, then, is the mechanism by which the hearer can come to tolerate such a wide range of irrelevant variation while remaining so selectively tuned to the critical features. This problem too is primarily one of concept formation. [SeePerception, article on Speech Perception.]
The young child’s experience with sequentially invariant routines can serve to induce perceptual equivalences that also resemble concepts. Whether the process is viewed as the classical conditioning of a terminal response to cues located earlier and earlier in the sequence, or whether it is viewed as the establishment of expectancies through chained anticipatory responses, the successive cues in such a routine all come to elicit a similar response pattern and are equivalent at least in the sense that any one of them will elicit the same response once the sequence is learned.
The common observation of sequential generalization in infants with respect to the mother supports this notion. At first only the touch of the nipple may elicit cessation of crying and reaching for contact. With repetition, however, earlier events in the nursing sequence, such as being picked up and held in a certain way, the sound of the mother’s voice, and the sound of the mother’s approaching footsteps or the opening of a door come to elicit the same response or an anticipatory portion of it. Typically the acquisition of such sequential equivalences of stimuli is analyzed according to a model of classical conditioning or operant chaining, but to the extent that a number of quite different cues come to have the same significance for the child, one could defend the view that concept formation of a primitive sort is taking place and that equivalence and expectancy may be the earliest examples of conceptual learning.
The perceptual capacities of the infant that involve possibly innate predispositions to differentiate and respond to a limited number of critical features of his environment have just come under study. Work on releaser stimuli, imprinting, and critical periods in infrahuman animals has helped to accelerate this mapping of prepotent discriminations in humans. Two outstanding examples are the work of Fantz on attention to patterned stimuli in the first six months of life and that of Gibson on avoidance of depth in the second half-year of life (see Gibson 1963). [SeeImprinting; Perception, article onDepth Perception.]
Exploration and recognition. A quite different research tradition in perceptual development is the Soviet literature on exploratory behavior and the recognition of objects, forms, patterns, and auditory pitch. Most of this work is untranslated, but a general summary has been provided by Pick (1963). The work is derived from a “motor-copy” theory of perceptual recognition, whereby incoming stimuli are matched against an image of the standard stimulus that is somehow stored in the memory of the organism. The most interesting feature of this approach is its experimental analysis of the image during initial training of children. An image is said to be formed and effectively stored to the extent that the child engages in voluntary scanning, handling, vocalizing, or other exploratory activity that provides him with feedback from this activity. The feedback must be isomorphic with critical features of the object being explored in order for a useful image to be formed. Thus the proprioceptive feedback from thorough, active exploration of relevant aspects of a stimulus is the stored image against which the feedback from subsequent exploration of new stimuli can be compared to determine equivalence or difference.
The significance of this approach is that at least potentially it provides a mechanism for the abstract coding and storage of knowledge that is neither symbolic nor verbal, unlike most models currently under investigation by Western psychologists, and is thus applicable to “preverbal” children. In implicit support of the motor-copy notion, there is good evidence that stereometric, three-dimensional objects which the preschool child can handle as well as look at produce more rapid and generalizable learning than do two-dimensional pattern stimuli or pictures, at which the child can only look.
Reading. Traditionally the borderline between perception and cognition has been obscure, and the same general assumptions and procedures have characterized some research in both areas. An outstanding recent example of the perceptual approach to cognitive development is the work of E. J. Gibson (1963) on children’s perception of letterlike forms, or graphemes. In Gibson’s work the child has to select the match of a standard artificial grapheme from an array containing a copy of the standard and a number of systematically constructed variants of the standard that are more or less difficult to discriminate from it. The transformations used to generate the array of variants are based on a prior analysis of critical and noncritical variations in letters of the alphabet. Examples of critical variations are line-to-curve transformations (e.g., D vs. O), break-to-close transformations (e.g., C vs. O), rotations and reversals (e.g., N vs. Z or W vs. M). Noncritical transformations included slants or tilts, with resulting compression, foreshortening, and other perspective distortions.
The results indicated that there are few errors and little improvement with age for break-to-close transformations; for the other transformations that were selected as critical for reading the Western alphabet, there were many errors at age four (nonreaders) but very few by age seven (readers). Discrimination on the basis of perspective transformations, which are not only noncritical for reading but actually represent variations which the child must learn to ignore in reading, showed the most errors of all, with little improvement from age four to age eight. [SeeLearning, article on Discrimination Learning; Reading Disabilities.]
These results, of course, have implications far beyond the development of grapheme perception and the process of learning to read. They confirm the prediction that as a child learns to discriminate perceptually on the basis of some differences and not others, he also acquires a general tendency to look at certain kinds of differences among any set of stimuli and to ignore others. The generality of such attentional sets and their ready application by the child to new stimulus materials are indeed impressive.
Observing and orienting responses
Just as the Soviet motor-copy theory of perceptual recognition demands the reproduction of critical features of the stimulus through relevant exploratory behavior, so a selective attention model of learning set formation requires that attention be viewed as a discriminative response that systematically selects certain features of a stimulus and makes them available for conditioning to a final response by selective reinforcement. (For a discussion of this approach, see Cantor 1965.)
General support for this point of view is provided in studies of learning under experimentally produced conditions of distraction or delay, where it can be demonstrated that distracting attention from the task interferes with learning more for younger than for older children, more in the early than in the later trials of the session, and more when the task is complex or ambiguous than when it is not.
While there are very few published studies wherein the observing responses of children have been directly manipulated or even systematically measured during learning, there is support from the literature on animal and human adult learning for the claim that relevant deployment of attention, as indicated by selective observing behavior, is itself a learnable response set that has considerable generality and power. It is not known, however, to what extent improvements with age in problem-solving ability can be directly traced to development of the capacity to attend to the most likely relevant cues at the outset of a problem, based on cumulative experience with similar problems.
The notion that attention can be adequately conceptualized as a set of centrally controlled, conditionable, voluntary, and discriminable observing responses is receiving continued emphasis in the research literature on learning in infrahuman organisms. At the level of overt behavior, activity such as vicarious trial and error in the rat, and reliance on the depth cue of motion-produced parallax in cats and chickens, provides evidence of the voluntary nature of important information intake processes. Similarly, physiological phenomena, such as the centrally mediated suppression of peripheral sensory activity (“peripheral gating”), are evidence of cortical control over the selection of cues, the acquisition of which should have massive effects on learning abilities. Additional evidence for the importance of observing behavior in children comes from experiments where stimulus presentation is deliberately designed to produce either interfering or facilitating effects with respect to relevant attending responses. Typically, corresponding effects on efficiency of learning are obtained. For example, in children’s discrimination learning, it appears to matter whether the locus of response is in the stimulus, immediately adjacent to it, or spatially removed from it. Similarly, reinforcement is most effective when its locus is closest to that of the stimulus and response and when a minimum of distracting consummatory behavior is elicited by delivery of reinforcement. [SeeAttention.]
These effects of relevant attending responses are most pronounced in the early stages of problem solving or learning and when the task is relatively difficult for the child. With the increasing familiarity and mastery that come with repetition and experience, the explicitly overt orienting and observing responses become attenuated and streamlined. Once a particular problem has been solved a number of times in different contexts, the elements require and elicit much less extensive and deliberate investigation, and the full-blown pattern of exploratory responses can then be elicited only by novel stimuli, unexpected outcomes, or arousal originating in some extrinsically motivating event, such as startle or threat. These phenomena of inhibition and disinhibition of investigatory behavior are discernible both as a function of chronological age and as a result of practice over relatively short intervals.
Just as spatial proximity of stimulus, response, and reinforcement facilitates learning by minimizing the occurrence of competing attending responses, so proximity in time contributes to learning efficiency by preventing distraction. Delay of response and delay of reinforcement are detrimental to learning, apparently as a result of the competition of irrelevant objects and events for the child’s attention during the delay interval, unless that interval is filled with relevant observing, labeling, or rehearsing behavior. If a child can be induced by pretraining, instructions, or the demands of the task to engage in some form of relevant thought during these delays, there is good evidence that the delay-produced deficit is replaced by an advantage in both acquisition and retention.
Distinct from the cortically controlled, “voluntary” observing responses are involuntary, autonomic orienting reactions that occur whenever attention is focused on a particular stimulus. These orienting reactions have been investigated most extensively by Soviet psychologists and physiologists, primarily by means of classical conditioning techniques. This work is important to cognitive development because of the significance of bringing attentional processes under experimental control and because much of this research has employed children as subjects. The catalogue of orienting reactions developed by Sokolov (1958) and others includes changes in the state of the sense organs, changes in the musculature that directs sense organs, changes in the skeletal-postural musculature, and changes in the central nervous system. Specifically, the pattern of orienting reactions that might be aroused by any attention-getting stimulus includes pupillary dilation; photochemical changes in the retina (which act to lower the absolute intensity threshold for light); turning of eyes, head, trunk, or whole body toward the source of stimulation; erection of the ears and sniffing in lower mammals; interruption of ongoing activity (especially proprioceptive behavior such as nonnutritive sucking); increase in general body tonus and rate of electromyographically recorded efferent impulses; desynchronization of slow-wave patterns in the electroencephalogram; constriction of blood vessels in the extremities, accompanied by dilation of those in the head; galvanic skin response (GSR); acceleration of heart rate; and momentary deceleration of respiration, followed by acceleration. Nearly all of these responses are subject to classical conditioning, and some are susceptible to operant control as well. To the extent that these attentional indicators are elicited in response to relevant cues with proper timing and sequence, they can facilitate the “higher mental processes,” such as conditional discrimination, concept formation, and rule-following behavior in complex tasks. [SeeNervous System; Senses.]
Semantic conditioning. Research on semantic conditioning (Razran 1961) suggests that the significance of a familiar stimulus to an individual child can sometimes be mapped in terms of the pattern of selective generalization of responses conditioned to that stimulus. Although the techniques employed are different, this work is directly comparable with American work on primary and secondary stimulus generalization in children. A typical procedure might begin by conditioning vasodilation or GSR by pairing presentations of the word “house” with mild electric shock or a loud noise of sudden onset and brief duration. Once conditioning is established, a test series of words is presented, including “house” but also, for example, “mouse” and “building.” A response to the former would be regarded as evidence for morphological (primary) generalization and is characteristic of young children. A response to the latter would constitute evidence for semantic (mediated or secondary) generalization and is characteristic of older children and adults.
Learning sets and transfer of training
Developmental studies of children’s learning have increasingly been concerned with the effects of certain kinds of pretraining experiences on subsequent learning and transfer test performances. To the extent that the task does not change substantially from pretraining to test, the experiment measures only recall or savings, and would not ordinarily be of much interest to the student of cognitive development. To the extent, however, that the pretraining establishes a general set, strategy, expectation, or principle that can be effectively transferred to a new set of materials, or be selectively applied to an entirely new task, then one could argue that the facilitating or interfering effects are germane to cognitive development. Similarly, a simple performance set or warm-up effect can be demonstrated in children as young as five years and can be experimentally differentiated from a discrimination learning set (learning to learn), of which children at this age are also capable (Cantor 1965). The extensive investigations of object discrimination learning sets in primates have been approximately replicated with positive results, using children as young as three years.
Studies comparing normal children with both retarded and gifted children of different ages have shown that with mental age (MA) held constant, learning sets are formed with increasing speed and precision as intelligence increases (as chronological age, or CA, decreases). (In these studies, intelligence is defined as synonymous with IQ.) Conversely, holding intelligence (MA/CA) constant, performance improves with increasing MA and CA. Pretraining of an appropriate attentional set facilitates performance in both normal and retarded children, while pretraining of an irrelevant attentional set produces interference when performance is compared with a control group receiving no pretraining. There is considerable evidence that learning difficulties in mentally retarded children can in large part be traced to a deficiency in acquiring a set to respond to relevant cues, and it has been proposed by a number of investigators that the key acquisition in a discrimination learning set is a general tendency to pay selective attention to relevant and informative cues. Zeaman and House (1963) demonstrated that the first improvement over a chance performance came on much later trials for retarded children than for normals of the same MA, though the rate of improvement once they started to improve was the same for the two groups. [SeeLearning, article on Transfer; Mental Retardation.]
The more generalized and abstract capacities and sets of which children are capable have been investigated most frequently either in the context of laboratory experiments on strategies of problem-solving and rule-following behavior or in a context of applied research on individual differences, curriculum development, and the acceleration of cognitive growth. Concept formation should be used as a rather inclusive generic term, denoting the most abstract and generalized aspects of grouping, ordering, and interrelating items of information. To the extent that a rule or principle abstracted from one experience or task or set of materials can be applied successfully to many others, it can be argued that a higher-order conceptual set has been established.
As early as 1941 Long and Welch demonstrated that young children could establish identity concepts more readily than relational concepts. Kuenne (1946) showed that while both school-age and preschool children could learn relational concepts, the range and scope of application to new stimuli was much greater among the older children. Heidbreder (1948) has established a general order of difficulty in concept formation; with some exceptions, such perceptually concrete features as object quality and shape are more readily learned as a basis for grouping than are derived attributes such as number.
More recently, developmental studies of groupings of common objects have demonstrated preference hierarchies for certain bases of classification (Bruner & Olver 1963; 1966). Between ages 5 and 11, the capacity to impose a superordinate conceptual structure on collections of terms develops, indicating an attempt by the child to achieve simplicity, economy, and parsimony of groupings without sacrificing specificity, discrimi-nability, and reproducibility of the members of a grouping on the basis of its definition. The bases of ordering terms begin with simple, functional groupings, equivalences by assertion, centration of associations to a single instance, chaining, and other asymmetrical combinations. With increasing age these groupings proceed toward greater ab-stractness, balance, and generality. [SeeConcept Formation.]
Other types of concept formation tasks, such as the Vygotsky blocks and the Wisconsin Card Sorting Test, have been used with children, but ordinarily in a context of clinical diagnosis or assessment of differential intellectual abilities and cognitive styles. Similarly, there is an extensive literature on the application of concepts, techniques, and materials originating in child psychology laboratories to training programs and curricula designed to foster more rapid conceptual growth (e.g., see Lovell 1961). Action research studies of real life environments have similarly taken as their point of departure the laboratory studies of the effects of stimulus enrichment and deprivation on learning and problem solving. Such efforts are typically aimed at defining, and in some cases improving, the potential stimulation toward abstract and conceptual thinking that is available in the child’s natural environment. [SeeMontessori.]
Language and verbal mediation
Perhaps the single most critical instrument of intellectual development is the acquisition of language, together with the capacity to use it as a working abstract representation of real objects, actions, and events. It would be difficult to overrate the significance of the development of working language, and more research has been focused on its relationship to complex learning and thinking than on all other aspects of cognitive growth combined. At the earliest and simplest level, words act as demands, first made on the child by others, then made by the child on his environment as well. The ontoge-netic origins of language, like its historical and cultural origins, are still subject to considerable debate, but there is general agreement that the emotional and expressive verbal behavior related to the child’s primary needs is the most primitive. The naming of objects is the next skill that emerges from the language of demands, as the child begins to use nouns to communicate his wants more specifically than is possible by the use of negative and affirmative or pleasure and distress indicators only. Naming things also appears to be the first significant step in the child’s use of language to help solve problems of a conceptual nature.
A major unsolved problem is, of course, how language as verbal behavior is learned. The conditioning approach has been taken in Soviet investigations of what Pavlov called the “second signaling system.” Starting with the assumption that perceptual responses to a real object are unconditioned, they have demonstrated that even in young children, successive presentations of a real object, immediately preceded by the sound of its name, result in the capacity of the name alone to evoke the “image” of the object by a process of ordinary classical conditioning. The next step was to demonstrate that second- and third-order conditioning could be established in children by using words as the conditioned stimuli to elicit verbal responses which in turn denoted the original object. Although many of the ordinary phenomena of classical conditioning, such as extinction, recovery, external inhibition, external disinhibition, and “irradiation” (generalization), can be demonstrated in this context, the power, generality, specificity, and flexibility of words as conditioning stimuli led Soviet psychologists to propose special laws for conditioning in the second signaling system, laws unique to the human organism.
There is, however, a growing tendency in the United States among psycholinguists to believe that linguistic competence is not the product of simple learning processes compounded and practiced over a number of years. Chomsky (1963), for example, believes that associative learning will not suffice to account for the acquisition of the structure of language, a structure that he contends is present in all languages and is apparent in the speech of surprisingly young children. While children doubtless imitate the expansions and corrections of their fragmentary speech that are provided by adults, there is evidence that even rudimentary exposure to grammatical language is often as effective as more structured kinds of linguistic training in eliciting grammatical speech. One wonders whether this reorientation may not require the assumption of a genetically determined and species-specific predisposition to structure speech. Such a predisposition would govern the structure and the motivation, while the environment would provide the idiomatic content of a particular language. The psychological model for such a process would resemble that of imprinting rather than that of classical conditioning or imitative operant learning. [SeeLanguage, article onLanguage Development; Learning, article OnVerbal Learning.]
From the behavioristic point of view, one unique aspect of words is the fact that they serve as both stimuli and responses. That is, the child can learn to manufacture verbal stimuli for himself almost as early as he can learn to respond to them as external events. Thus, there has been considerable research on how the child comes to use words, not only as abstract representations of objects not present but also as the basis for conditional self-instructions for the monitoring and guiding of complex sequences of goal-oriented, instrumental behavior.
By about the age of four the child is able to group objects, in part on the basis of their names. Investigators such as Spiker (1963) have shown that cues can acquire distinctiveness or similarity, depending upon the distinctiveness or similarity of the names attached to them and the strength of that attachment. The possession of similar-sounding names facilitates the grouping of disparate objects into the same class and interferes with learning to discriminate them into separate classes. Conversely, the acquisition of distinctive names for objects facilitates their discrimination and interferes with equivalence learning, to the extent that the names have a distinctive sound and are learned sufficiently thoroughly that they are not misapplied to their referents. By age six or seven, the child begins to use language not just to code stimuli but to mediate new learning and thinking.
Mediation theory. Mediation theory, specifically verbal mediation theory, has developed in the United States within the body of stimulus-response behaviorism. The key concept of mediation theory is the intervening response and the distinctive feedback that regularly result from its occurrence. Any response will qualify as such a mediator if it can be conditioned either classically or instrumen-tally to an initial stimulus and if it produces reliable proprioceptive stimulation to which a final motor response can, in turn, be conditioned. The original prototype of the intervening response was Hull’s fractional anticipatory goal response (rG). Such responses were said to produce feedback stimulation that strengthened running behavior in the rat by virtue of the fact that rG was selectively conditioned to external stimuli along the previously reinforced route to the goal box. The notion was adopted and generalized by Osgood (1953) to suggest that if verbal or autonomic responses were conditioned to any stimulus, the proprioceptive feedback from these responses could serve to define the meaning of the stimulus in an associational context. If the name of an object is the initial stimulus, the same mediating chain of intervening response and feedback will operate in the absence of the object itself, leading to the same behavior as when the object itself was present as the initial stimulus. When the intervening response is a verbal mediator (either overt or covert), the response-produced stimulation functions in the same way as the response, and the possibility of self-instruction is literally realized.
The general intervening response model can encompass not only the effects of mediation on verbal learning and verbal behavior but phenomena such as acquired distinctiveness or equivalence of cues, relational learning, and inferential learning as well. Broadly interpreted as a feedback model, it is also applicable to semantic conditioning, selective observing behavior, learning sets for relevance of cues, and even a motor-copy theory of perceptual recognition. In addition, it provides the child with a degree of abstract and flexible control over his own behavior that is compatible with behavioristic determinism yet consonant with the rather high degree of purpose, insight, and intelligence that is commonly inferred from observation of children’s voluntary behavior. And it does so by means of a rather simple mechanism that would have been branded as unacceptably teleological by many of the behaviorists of the 1940s. Intervening response and feedback models, especially as applied to verbal processes, are among the most promising for contemporary research and future understanding of cognitive development. These models are theoretically neutral with respect to the problem of language acquisition but delineate two tasks that are prerequisites for language to facilitate thinking : the production of relevant verbal behavior and the use of the resulting feedback to control conceptual behavior. Finally, they are compatible with the internal logic of computer-based models for the simulation of human cognitive processes. [SeeSimulation, article onIndividual Behavior.]
Thus language serves to keep the child on the track of his own intentions, preventing irrelevant detours, ordering his responses in proper sequence, and informing him whether his behavior is bringing him closer to the desired conclusion. In the course of verbalizing a plan for solving a problem and developing self-instructions for monitoring progress toward solution, the child is increasingly able to codify whole sets of information-processing rules and strategies with sufficient generality to give him positive transfer from one problem to the next, even when the content is radically different. Not only can he develop a repertoire of learning sets that facilitate solutions to new problems when they are logically similar to those previously mastered, but he can also label these sets, strategies, or problem-solving routines themselves. Thus they can be called upon when appropriate, much as a computer stores and identifies subroutines to be called whenever the program or its output requires them (see Miller et al. 1960).
Beyond the study of stages and mechanisms of cognitive development, and beyond the testing of general intellectual potential by standard means, there is a considerable body of research on qualitative and stylistic differences in cognitive functioning among individuals. It has long been recognized that certain aspects of personality and temperament can both influence and be influenced by cognitive capacities, but systematic and objective research into the structure of cognitive styles and their influence on thinking and subsidiary processes is fairly recent. The same is true of motivational factors inherent in cognitive activity. General personality variables that have been shown to be correlated with cognitive performances include achievement motivation, authoritarianism, manifest anxiety level, somatotype, and “ego control,” to name but a few. Typically, such research has shown how a well-organized characterological predisposition can have specific biasing effects on thinking, reasoning, and problem solving. In most cases such effects have not been studied develop-mentally, and although there is considerable speculation, little hard knowledge is available on the ontogenesis of such traits. Other stylistic variables have emerged directly from research on perceptual, learning, and cognitive processes themselves rather than from personality theory. Such variables as field articulation (Witkin et al. 1954; 1962), reflection-impulsivity (Kagan et al. 1964), and creativity (Getzels & Jackson 1962; Wallach & Kogan 1965; Guilford 1956) are beginning to receive the research attention they deserve.
As is the case in personality research, the naming of syndromes and the demonstration by correlational means that certain clusters of traits tend to inhere in the same individuals do not explain the development and causation of such patterns. Nor do they suffice to demonstrate that similar patterns often characterize parents and their children, attributing the perpetuation of styles to genetic factors, imitative learning, or selective reinforcement. While such effects undoubtedly do occur, more functional study of the genesis of cognitive styles in individuals is sorely needed. One successful example is the work on origins of need for achievement in early independence training of children. Another is the quasi-psychoanalytic account of the development of intolerance for ambiguity as an aspect of the authoritarian personality. [SeeAchievement Motivation; Personality, Political, article onConservatism AND Radicalism.]
Without excessive oversimplification, it is possible to regroup cognitive style variables into two general classifications: one dealing with the precision, accuracy, stability, and independence of cognitive processes; the other dealing with their fluency, spontaneity, and openness. For example, under precision one might include rigidity, reflectivity, field articulation and independence, ego overcontrol, and intolerance of ambiguity, among others. Under fluency one could group flexibility, originality, creativity, impulsivity, field dependency, and ego undercontrol. Depending upon the variables included, these two classifications would be either independent or negatively related. The number of orthogonal (independent) syndromes to be defined appears to be strikingly limited, and the relations among a plethora of proposed variables, clusters, and factors have yet to be satisfactorily resolved.
Even when such a resolution is achieved, the resulting syndromes need careful developmental study. Two variables whose origins and incidence in children have been investigated are creativity and reflectivity. Research on creativity has been carried on independently in a number of research centers with different orientations, objectives, and methods. The current status may be summarized as follows: A loose cluster of variables called creativity can be reliably distinguished from intelligence by a number of objective means. A tendency to produce novel, clever, unusually organized, original, or simply uncommon responses in a variety of tasks, while it is often a consistent trait in the individual, does not predict general intellectual accomplishment as well as intelligence test scores do. As a stylistic variable it seems to develop most commonly in a generally flexible, stimulating, casual, nonconforming, and permissive home environment, but no particular set of experiences has been identified as the major causal determinant. It does not correlate with social and emotional adjustment any more highly than it does with intelligence. [SeeCreativity, article on Psychological Aspects.]
The style variable of re-flection-impulsivity has been investigated by Kagan and his collaborators (Kagan et al. 1963; 1964) and is a likely candidate to account for a good deal of individual variability in perceptual, conceptual, and motor tasks involving speed and accuracy. It is of interest to students of intellectual development because it has been investigated extensively in children, as well as in adults, and because it correlates consistently with an extensive cluster of performance measures. A number of different tasks have been developed to serve as the criterion measure of reflection, or analytic style, as it has sometimes been called, but all of them have three features in common: (1) The tasks contain some degree of ambiguity or response uncertainty, such that no response is obviously correct at the outset. Thus it is possible that in assessing how people tackle a cognitive task they are in fact measuring some mixture of preference and ability. (2) The criterion tasks require a certain amount of perceptual analysis of detailed, often minute stimulus differences. They require careful, orderly comparisons among and between stimuli in order to yield a highly reflective score, but they are in no sense tests of sensory acuity. (3) Either directly or indirectly they take account of the speed of response as well as the nature of response, the faster performances generally being classified as impulsive.
Performances at the reflective, analytic end of this style dimension are in general precise, relatively slow, systematic, logical, objective, and somewhat literal-minded. Reflective adults in the Fels Institute longitudinal sample displayed, as young children, less physical activity, motihty, and restlessness; longer attention spans; and less dis-tractibility than did those adults who scored at the impulsive and global end of the distribution. The impulsive, nonanalytic syndrome, conversely, is characterized by fast, relatively inaccurate, and inconsistent, though perhaps more imaginative, performance in perceptual-cognitive tasks. One recent study (Kagan 1965) has given some indication of a relationship between impulsivity and reading difficulties in the primary grades. There is, moreover, evidence of sufficient stability of this characteristic from as early as the second year of life to warrant consideration of the possible contribution of genetic determinants. A developmental trend of declining impulsivity and increasing reflectivity with age has also been noted. Like other promising style variables, reflectivity possesses both a developmental and a differential component and will require both kinds of analysis if we are fully to understand its epigenesis and its role as a modifier of individual intellectual development.
John C. Wright
[Directly related are the entriesDevelopmental Psychology; Intelligence And Intelligence Testing; Language, article onLanguage Development; Moral Development; Perception, article onPerceptual Development; Sensory And Motor Development. Other relevant material may be found inAdolescence; Educational Psychology; Infancy; Learning, articles onClassical Conditioning, Instrumental Learning, Discrimination Learning, Verbal Learning, LearningIn Children, Transfer; Problem Solving; Reasoning And Logic; and in the biography ofMontessori.]
Bruner, Jerome S.; and Olver, Rose R. 1963 Development of Equivalence Transformations in Children. Society for Research in Child Development, Monographs 28, no. 2:125-141.
Bruner, Jerome S.; and Olver, Rose R. (editors) 1966 Studies in Cognitive Growth. New York: Wiley.
Cantor, Joan 1965 Transfer of Stimulus Pretraining to Motor Paired-associate and Discrimination Learning Tasks. Volume 2, pages 19-58 in Advances in Child Development and Behavior. Edited by Lewis P. Lipsitt and Charles C. Spiker. New York: Academic Press.
Chomsky, Noam 1963 Formal Properties of Grammars. Volume 2, pages 323-418 In R. Duncan Luce, Robert R. Bush, and Eugene Galanter (editors), Handbook of Mathematical Psychology. New York: Wiley.
Darwin, Charles 1877 A Biographical Sketch of an Infant. Mind 2:285-294.
Falkner, Frank (editor) 1966 Human Development. Unpublished manuscript. → See especially the article on cognitive development.
Getzels, Jacob W.; and Jackson, Philip W. 1962 Creativity and Intelligence. New York: Wiley.
Gibson, Eleanor J. 1963 Perceptual Development. Pages 144-195 in National Society for the Study of Education, Child Psychology. 62d Yearbook, Part 1. Edited by Harold W. Stevenson. Univ. of Chicago Press.
Guilford, J. P. 1956 The Structure of Intellect. Psychological Bulletin 53:267-293.
Heidbreder, Edna 1948 The Attainment of Concepts: VI. Exploratory Experiments on Conceptualization at Perceptual Levels. Journal of Psychology 26:193-216.
Kagan, Jerome S. 1965 Reflection-Impulsivity and Reading Ability in Primary Grade Children. Child Development 36:609-628.
Kagan, Jerome S.; Moss, Howard; and Sigel, Irving 1963 Psychological Significance of Styles of Conceptualization. Society for Research in Child Development, Monographs 28, no. 2:73-112.
Kagan, Jerome S. et al. 1964 Information Processing in the Child: Significance of Analytic and Reflective Attitudes. Psychological Monographs 78, no. 1.
Kendler, Tracy S. 1963 Development of Mediating Responses in Children. Society for Research in Child Development, Monographs 28, no. 2:33-48.
Kuenne, Margaret R. 1946 Experimental Investigation of the Relation of Language to Transposition Behavior in Young Children. Journal of Experimental Psychology 36:471-490.
Long, Louis; and Welch, Livingston 1941 Reasoning Ability in Young Children. Journal of Psychology 12: 21-44.
Lovell, Kenneth 1961 The Growth of Basic Mathematical and Scientific Concepts in Children. New York: Philosophical Library.
Luriia, Aleksandr R. 1961 The Role of Speech in the Regulation of Normal and Abnormal Behavior. New York: Pergamon; Liveright.
Miller, George A.; Galanter, E.; and Pribram, K. H. 1960 Plans and the Structure of Behavior. New York: Holt.
Montessori, Maria (1909) 1964 The Montessori Method: Scientific Pedagogy as Applied to Child Education in “The Children’s Houses.” Cambridge, Mass.: Bentley. → First published as 11 metodo della peda-gogia scientifica.
Osgood, Charles E. (1953) 1959 Method and Theory in Experimental Psychology. New York: Oxford Univ. Press.
Piaget, Jean; Vinh-bang; and Matalon, B. 1958 Note on the Law of the Temporal Maximum of Some Optico-Geometric Illusions. American Journal of Psychology 71:277-282.
Pick, Herbert L. Jr. 1963 Some Soviet Research on Learning and Perception in Children. Society for Research in Child Development, Monographs 28, no. 2: 185-190.
Pick, Herbert L. Jr.; Pick, Anne; and Klein, Robert 1966 Perceptual Integration in Children. Unpublished manuscript.
Razran, Gregory 1961 The Observable Unconscious and the Inferable Conscious in Current Soviet Psycho-physiology : Interoceptive Conditioning, Semantic Conditioning, and the Orienting Reflex. Psychological Review 68:81-147.
Rheingold, Harriet L.; Gewirtz, Jacob L.; and Ross, Helen W. 1959 Social Conditioning of Vocalizations in the Infant. Journal of Comparative and Physiological Psychology 52:68-73.
Sokolov, Evgenii N. (1958) 1963 Perception and the Conditioned Reflex. New York: Macmillan. → First published as Vospriiatie i uslovnyi refleks.
Spiker, Charles C. 1963 Verbal Factors in the Discrimination Learning of Children. Society for Research in Child Development, Monographs 28, no. 2:53-69.
Wallach, Michael A.; and Kogan, Nathan 1965 Modes of Thinking in Young Children. New York: Holt.
Werner, Heinz (1926) 1957 Comparative Psychology of Mental Development. Rev. ed. New York: International Universities Press. → First translated from German in 1940. The revised edition was first published in 1948.
White, Sheldon H. 1965 Evidence for a Hierarchical Arrangement of Learning Processes. Volume 2, pages 187-220 in Advances in Child Development and Behavior. Edited by Lewis P. Lipsitt and Charles C. Spiker. New York: Academic Press.
Witkin, Herman A. et al. 1954 Personality Through Perception. New York: Harper.
Witkin, Herman A. et al. 1962 Psychological Differentiation: Studies of Development. New York: Wiley.
Wohlwill, Joachim F. 1960 Developmental Studies of Perception. Psychological Bulletin 57:249-288.
Zeaman, David; and House, Betty J. 1963 The Role of Attention in Retardate Discrimination Learning. Pages 159-223 in Norman R. Ellis (editor), Handbook of Mental Deficiency. New York: McGraw-Hill.