PSYC 525 Developmental Psychology

Developmental Psychology 1992, Vol. 28, No. 3, 360-367

Copyright 1992 by the American Psychological Association, Inc. OOL2-1649/92/$3.0O

John B. Watson's Legacy: Learning and Environment

Frances Degen Horowitz Graduate School and University Center

City University of New \brk

John B. Watson's contribution is evaluated in relation to his own time, with respect to his historical influence, and in light of current issues in developmental psychology. A survey of a nonrandom sample of current developmental psychologists revealed no consensus with respect to Watson's legacy to developmental psychology. The influence of Watson's insistence on an objective methodol- ogy in psychology remains, although is not necessarily acknowledged. His extreme environmental- ism has been rejected. His concern to understand the principles of learning is reflected in the subsequent work of the Hullians and the Skinnerians. The influence of his underlying premise about the importance of environment and of learning is to be found in such work as studies of the effects of intervention programs. I question the possible costs to the field of the continued rejection of a Watsonian emphasis on learning as an important process in development and behavior.

In his time, John Broadus Watson (1878-1958) was contro- versial. So he remains. In preparation for this reflective essay on his legacy to developmental psychology, I sent a question- naire to a nonrandom sample of mostly senior developmental psychologists.1 The 45 respondents ranged widely in their per- ceptions of Watson's contributions and in their evaluations. Watson's name, it appears, can still elicit strongly negative reac- tions. For example, in relation to his contribution to develop- mental psychology and psychology in general, Watson was de- scribed as "an embarrassment," "harmful," "very important, but mostly a negative influence." One respondent said his "main contribution was obfuscation," another that "he had lit- tle lasting effect," and yet another that the "long-term effect of Watson was harmful."

Other comments were more positively valenced, such as, that his "biggest contribution was balancing the scales between na- ture and nurture by overstating the case," his "major contribu- tion was to establish behavior as an important phenomenon in its own right," his "thoroughgoing empirical orientation was of the greatest import," and his "methodological behaviorism is to be found in neobehaviorism, behavior analysis and much of cognitive psychology." One person wrote, "Contemporary developmental psychology would not be the same had not Wat- son contributed to this field."

John B. Watson died more than 40 years ago. His active con- tributions to the field of psychology ceased more than 50 years ago, and an evaluation of him and his work still evokes strong reactions and little consensus about his legacy. About half of the respondents rated Watson on the positive side in agreeing that his main contribution was in the area of methodology; the other half shaded to the disagreeing side. A few more than half

The original draft of this article was written while the author was at the University of Kansas.

Correspondence concerning this article should be addressed to Frances Degen Horowitz, President, Graduate School and University Center, City University of New York, 33 West 42nd Street, New York, New York 10036-8099.

felt he had ignored biological variables, but the remainder tended to disagree with that characterization.

Diversity of opinion is rife in developmental psychology to- day, but when asked to evaluate the theoretical and methodolog- ical contributions of Baldwin, Binet, Darwin, Freud, Gesell, Hall, Lewin, Piaget, and Vygotsky, this same group of respon- dents showed much more consensus. This was true even for the figures in the list whose active careers had ceased around the same time as or before Watson's. For example, the theoretical contributions were strongly and positively acknowledged for Darwin, Freud, Lewin, Piaget, and Vygotsky. There was also a general consensus about the methodological contributions of Binet, Lewin, and Piaget.

Evaluating John B. Watson's legacy to developmental psychol- ogy thus appears to involve a considerably more complex set of considerations than is the case for other major figures. Indeed, some of the respondents objected to identifying Watson as a developmentalist, claiming that he was, at best, a psychologist concerned only with defining psychology as a natural science and, at worst, a dogmatist who went far beyond his data to popularize his beliefs about development. If there was any con- sensus to be found, it was that Watson's tireless championing of behaviorism as the only acceptable way of looking at behavior succeeded in making his point of view the dominant one for many years.

To be sure, there were balanced evaluations: Some of them were thoughtful and lengthy, pointing out that Watson had to be

11 wish to acknowledge the following people, who responded to the survey: Donald Baer, Harry Beilin, Sidney Bijou, Andy Collins, John Colombo, Cynthia Deutsch, John Flavell, Norman Garmezy, Roberta Golinkoff, Charlie Greenbaum, John Hagen, Willard Hartup, Aletha Huston, Dick Jessor, Jerome Kagen, William Kessen, Claire Kopp, Michael Lamb, Jonas Langer, Lewis Lipsitt, Bob McCall, Neal Miller, Edward Morris, David Palermo, Anne Pick, Herb Pick, Robert Siegler, Hayne Reese, Carolyn Rovee-Collier, Arnold Sameroff, Irving Sigel, Charles Spiker, Joe Spradlin, Harold Stevenson, Sheldon White, Mont Wolf, John Wright, plus eight other respondents, who chose not to provide their name.

360

APA CENTENNIAL: JOHN B. WATSON'S LEGACY 361

evaluated in the context of his time, that his extreme views were necessary to establish behaviorism, and that the field was the better for it in the long run even as less dogmatic perspectives came to pass. There was also consensus, among those who chose to comment on it, that the extreme nature of his environ- mentalism as it was translated into popular child rearing advice was regrettable.

It will be useful, in examining Watson's legacy to developmen- tal psychology, to consider his contributions in his own time, his influence in the years after he ceased to be a contributor, and his current and perhaps future impact.

In His Own Time

The opening sentences of Watson's (1913) manifesto, "Psy- chology as the Behaviorist Views It," declare clearly and un- equivocally that "psychology as a behaviorist views it is a purely objective experimental branch of natural science. Its theoretical goal is the prediction and control of behavior" (Watson, 1913, p. 158). He was, as everyone notes, sounding the charge against introspection, establishing a purely American position in the field, and, interestingly enough, in a somewhat anti-Darwinian spirit, suggesting there was no necessary relationship between the laws governing animal behavior and human behavior.

Seven years later Watson was deeply embroiled in the nature- nurture controversy. He had claimed the principles of learning as the central and practically only variable controlling the ac- quisition of human behavior. He had come to focus his atten- tion largely on the application of behaviorism to the study of the behavior of infants and young children. And he was soon to leave his academic position at Johns Hopkins University and take up employment in the field of advertising even though he would continue to write about behaviorism and to engage in the popularization of his ideas in magazines and on the radio for almost 20 years (Buckley, 1989; Nance, 1970).

It is not easy to separate the impact of Watson's ideas about behavior and development in his own time from Watson's per- sona. By all accounts he was a man given to opportunism, to making extreme statements, to evincing strong ego needs for visibility and notoriety, and to displaying a cold and imperious style. He did not have many personal supporters when he was ousted from Johns Hopkins University in the wake of what was then considered public scandal in relation to his divorce and remarriage (Buckley, 1989). It is not clear whether any univer- sity would hire him after that, and he went, quite profitably, into the field of advertising, where his application of behavior- istic principles was very successful.

Although the 1913 statement defining the behavioristic view of psychology was not about development, Watson's major focus, ultimately, was on the development of behavior. Even so, many developmentalists today do not count him as a develop- mentalist. In his time, neither Watson nor most of his contem- porary developmentalists distinguished between behavioral de- velopment and the acquisition of specific responses. Gesellians saw behavioral acquisition and development as an unfolding of largely inherited behaviors. Watson—taking an opposing, envi- ronmental point of view—credited learning, namely condi- tioning, as the sole process responsible for development.

By the time Watson published the first edition of his text,

Psychology From the Standpoint of a Behaviorist in 1919 (which had second and third revised editions in 1924 and 1929, respec- tively), he was applying behavioristic analyses to development and using the analyses to explain the acquisition of behavior. The basic principles he asserted in the 1919 publication were repeated in his 1924 publication Behaviorism (Watson, 1924/ 1970) and are relevant to my discussion. First, Watson exhib- ited a healthy respect for the biological functions of the human organism, discussing at length what was then known about genes, the nervous system, and the human muscular system. He recognized the human body as an extremely complex organ- ismic system that was highly integrated with the behavioral system. He was wont to stress, again and again, that the body acts as a whole and that behavior is rooted in and roots the organism.

The second observation is that Watson clearly believed that learning was almost entirely responsible for behavioral develop- ment but also acknowledged the role of structural change. In fact, in discussing the motor behaviors that develop in the in- fant and young child, he used italics to stress the point:

In the great majority of these later activities [i.e., crawling, standing, sitting up, walking, running, jumping] it is difficult to say how much of the act as a whole is due to training or conditioning. A considerable part is unquestionably due to the growth changes in structure, and the remainder is due, we believe, to training or condi- tioning. (Watson, 1924/1970, p. 135-136)

A third observation involves the strategies and program Wat- son advocated for developmental research. He stressed the need to study infants, he focused heavily on the central role of emo- tions, and most obviously, he was relentless in his insistence on learning as the major mechanism for explaining behavior and development. He had already used the experimental method to show learning as the basis for the acquisition of emotional re- sponses in the young infant. The particular experiment for which Watson is most known purported to demonstrate how emotional reactions in a young child could be conditioned and to suggest, by implication, that this was the model for the ac- quisition of most emotional responses (Watson & Rayner, 1920). This experimental effort, involving the use of infants to study emotional behavior, was considered a breakthrough, though there had been earlier reports of the conditioning of motor responses using infants (Krasnogorski, 1909; Mateer, 1918). Its findings and advocacy for the experimental strategy were strengthened in subsequent reports by Jones of experi- ments designed to show the conditioning and the uncondition- ing of the fear response in the very young child (Jones, 1924a, 1924b).

Watson's developmental model was exceedingly simple, con- taining no discussion of stages and little of sequences; there was no consideration that learning principles were in any way in- fluenced by the age of the child. Furthermore, the developmen- tal progression, despite the nod to structural change as a vari- able, was linear and cumulative.

Finally, Behaviorism (Watson, 1924/1970) reflects the major thrust of Watson's position, the aspect that served as the main lightning rod for his critics and caricaturists then and now: a practically unqualified belief in the role of experience and envi- ronment in shaping the human behavioral repertoire.

362 FRANCES DEGEN HOROWITZ

Give me a dozen healthy infants, well-formed, and my own speci- fied world to bring them up in and I'll guarantee to take any one at random and train him to become any type of specialist I might select—doctor, lawyer, artist, merchant-chief and, yes, even beg- gar-man and thief, regardless of his talents, penchants, tenden- cies, abilities, vocations and race of his ancestors. (Watson, 1924/ 1970, p. 104)

This passage is widely quoted and often ridiculed, but it is generally provided without the two sentences that follow it:

I am going beyond my facts and I admit it, but so have the advo- cates of the contrary and they have been doing it for many thou- sands of years. Please note that when this experiment is made I am to be allowed to specify the way the children are to be brought up and the type of world they have to live in. (Watson, 1924/1970, p. 104)

Throughout the book, Watson the scientist took care to indi- cate the limitations of the data and the need for more research. After presenting a strong argument for the pervasiveness of conditioning of emotional reactions, he concluded with a cau- tionary note drawing the reader's attention to the fact that all his conclusions "are based now upon too few cases and too few experiments" (Watson, 1924/1970, p. 195), but he was optimis- tic that those already at work on the problems would rectify the situation.

Watson was not only scientist but advocate, in this book and elsewhere. The aspect of Watson's career that elicits the stron- gest negative reactions was his willing and unqualified popular- ization of his belief in the efficacy of extreme measures of envi- ronmental control in the form of advice to parents and teachers about the rearing of children. For example, he took the position that physical affection and expressions of love impeded good development. His judgments about what was good and bad for children rested on a model in which he advised that it was good to limit the influence of emotion on behavior and to train chil- dren so as to maximize independence and skill acquisition. Interestingly, he appeared to accept Freud's observations, made reference to Freudian concepts, and believed them entirely ame- nable to a learning analysis.

By any standard his environmentalism was unbridled. He saw no role for inherited characteristics as ultimately having any determining role in developmental outcome. He did not deny genetic influences, but he believed that what the environ- ment provided in the way of experience and training could override organismic variables and ultimately determine devel- opmental outcome. Watson's view about human potential was egalitarian in the extreme, and he plied the perspective in every arena.

Watson did not shy away from admitting that there are in- herited differences, but he made a distinction between the in- heritance of structures and the inheritance of function. The former was clearly heritable, the latter not. Function was the result of how the environment shaped the hereditary structure, such shaping beginning in the prenatal environment. He re- jected behavioral differences as being due to racial or other hereditary variables, claiming that environmental experience alone would account for observed differences (Watson, 1924/ 1970). Watson, on every issue, took the extreme environmental position in the nature-nurture controversy.

A tireless advocate for the relevance of psychology, Watson

believed strongly in the application of psychological principles to the solution of practical problems. He plumped for the im- portance of applied research; his view of its potential for doing good in society was optimistic in the extreme. In some of this he was not unlike Gesell. Perhaps it was part of the American Zeitgeist that Gesell and Watson both believed that the applica- tion of the principles of development (in Watson's case the prin- ciples of learning) to the rearing of children would result in happier children, despite GeselPs not buying any of Watson's environmental determination of developmental outcome.

The America of Watson and of Gesell was coming to value science more and more, to see science as a panacea for social ills. Industrialization had taken hold. Science applied to mak- ing an industrialized society efficient and socially progressive was the desideratum. Watson's ideas about social engineering were a positive fit for the times.

Watson Diffused

By the end of the 1930s, John B. Watson had been separated from academic psychology and laboratory science for almost 15 years. His expression of his own point of view remained static, unperturbed by new data or theoretical advance. His active participation in psychology had begun to decline, but his meth- odological behavioristic credo that insisted on the necessity for objectively collected, independently verifiable data gained wide acceptance and became the standard for doing psychologi- cal science.

Though Watson's persona was beginning to fade from the field, his influence spread to experimental psychology, to social psychology, and to developmental psychology. Clark Hull and his Yale colleagues and students took up the interest in condi- tioning. They elaborated on Watson and Pavlov and explored the parameters of classical and instrumental conditioning. They embellished the philosophical rationale for behaviorism by adopting the point of view of the logical positivists; they made formal theory drive experimental research in a manner exceedingly more sophisticated than Watson's efforts (Hull, 1943). One difference between the Hullians and Watson was that the Hullians were more inclined than Watson to see animal behavior as an analogue for simple human behavior.

A Watsonian emphasis on stimulus-response (S-R) relations and the primacy of learning in the context of a Hullian ap- proach found outlets in work on imitation (Miller & Dollard. 1941), frustration and aggression (Miller, 1941), and personal- ity and psychopathology (Dollard & Miller, 1950). Watson's be- lief in the efficacy of studying the young, relatively naive organ- ism to understand how the principles of learning operated was reflected in the flourishing field of experimental child psychol- ogy (McCandless & Spiker, 1956).

The explicit developmental emphasis in all of these efforts was mainly muted. There was not much developmental theory per se guiding the research. A linear, cumulative model was generally accepted, if not articulated, though an early study by Kuenne (1946) introduced something of a counterweight. Kuenne% work on transposition suggested that the level of a child's language development modified how learning occurred, thus introducing a quite modest developmental caveat into the learning literature. A stronger developmental focus was eventu-

APA CENTENNIAL: JOHN B. WATSON'S LEGACY 363

ally to find expression in the work of the Kendlers (H. H. Kendler & Kendler, 1961,1969; T. S. Kendler & Kendler, 1959; 1966; Kendler, Kendler, & Learnard, 1962).

The translation of Freudian theory into behavioristic terms and an emphasis on dyadic interactions stimulated what eventu- ally was to become social learning theory (Sears, 1951). A stimu- lus-response analysis married to Freudian developmental stages provided for a strong behavioristically oriented develop- mental approach with ties back to Watson through the Miller- Dollard work. During the 1950s Watson's methodological influ- ence was unquestioned. His methodological influence on devel- opmental and child psychology research was pervasive. So, to a lesser degree, was his environmentalism.

Watson's most direct descendent is generally identified as B. E Skinner. Skinner published The Behavior of Organisms in 1938, focusing on operant learning as the basic mechanism controlling behavioral acquisition. The influence of this book may have been restricted by the subsequent onset of World War II and by the diversion of the efforts of most psychologists, Skinner included, to the war effort. An alternate view is that Skinner's effort would, in any event, have been eclipsed by the intensity of the debate between Hull and Tolman and the advo- cates for their positions.2

Skinner's (1953) publication of Science and Human Behavior restated many of the basic views of the 1938 book, though it was less theoretical and had a more applied focus. A reading of Science and Human Behavior gives one a strong sense of being in contact with "Watson updated" and "Watson sophisticated," although Skinner's emphasis on the overriding importance of contingent reinforcement in the shaping of human behavior might well have been deemed too teleological by Watson. The empirical work stimulated by Skinner was fully in the Watson- ian methodological tradition and revived the labeling of these efforts as radical behaviorism, with radical suggesting ex- tremism.

Watson and Skinner shared an unyielding commitment to environmentalism, though Skinner's analysis was placed in a larger evolutionary context than Watson ever entertained (Skinner, 1974). Skinner advocated an almost singular focus on operant conditioning; Watson recognized both instrumental and classical conditioning. Watson's interest in classical condi- tioning stemmed, in part, from his conviction about the central- ity of emotional behavior. He felt, from an applied point of view, that it was important to be very careful about how the emotions were conditioned early in life so as to shape an inde- pendent, self-reliant, and unemotional personality that would thus result in the most functional and happy of people.

The application of the Skinnerian approach to study the ba- sic principles of learning in children was taken up by Bijou (1955,1957,1958) and Bijou and Baer (1961,1963). The basic research on children's learning continued, but it was the growth of applied research under the rubrics of behavior modification or behavior analysis that came to characterize the Skinnerians. Watson's vision for an applied psychological science driven by experimental results was to find its fullest realization in the work inspired by Skinnerian principles. Skinner, himself, had served the theoretical vision in writing Walden Two (Skinner, 1948).

Watson would have cheered the outpouring of research de-

signed to apply behaviorism to the real world, research con- ducted first by Skinnerians and subsequently in fields such as behavioral medicine, industrial and organizational psychology, and community psychology. Particularly, Watson would have applauded the testing and application of the principles of learn- ing to improve the functioning of persons with mental and physical handicaps, to improve classroom deportment and learning, and to modify the behavior of delinquent youths.

In 1970 the Journal of Applied Behavioral Analysis was founded. It lent the field of applied behaviorism a professional stamp. A year before, in his essay on "Behaviorism at Fifty," Skinner declared, "In the fifty years since a behavioristic philo- sophy was first stated, facts and principles bearing on the basic issues have steadily accumulated" (Skinner, 1969, p. 228). Wat- son would have, with gratification, agreed.

Watson's Influence Now and in the Future

Gustav Bergmann (1956) considered that John Broadus Wat- son was, with the exception of Freud, "the most important figure in the history of psychological thought during the first half of the century" (p. 265). Skinner's obituary for Watson, appearing in Science, likened Watson's stature to that of Dar- win and Lloyd Morgan (Skinner, 1959). Bergmann's estimate rested solely on Watson's methodological contribution of wrest- ing psychology from introspection and mentalisms. He deemed Watson's social philosophy and what he called Watson's meta- physical outlook as "silly" (Bergmann, 1956). Today Skinner is widely—and somewhat erroneously—regarded as being as arch an environmentalist as Watson was; Skinner's estimate of Wat- son, however, was that his extreme environmentalism and his inclination for being polemical undermined both his impact and his effectiveness.

One does not know, of course, how and to what degree Wat- son's ideas would have changed had he lived the rest of his life as an active empirical psychologist or, consequently, what impact he might have had on the field of developmental psychology. It is something of a paradox that in the 1950s Bergmann and Skinner placed Watson in the same league as Darwin and Freud, whereas some psychologists today regard Watson as an embarrassment and as having done harm to the field. Some today make the quite harsh judgment that Watson cost psychol- ogy in general, and developmental psychology in particular, 50 years of floundering, using a wrong and unproductive para- digm. Others still see, on balance, Watson's legacy as positive and enduring. Some of the distinctions that he insisted on mak- ing with respect to definitional and methodological practices, for example, have remained cornerstones in psychology. Definitional standards for stimuli and responses and criteria for making objective and reliable observations can be traced back to Watson and continue to characterize acceptable investi- gatory practices today.

To weigh these contradictory estimates of Watson's contribu- tion, it is necessary to consider the current condition of develop- mental psychology, particularly as practiced in the United

21 am indebted to the anonymous reviewer for this alternative per- spective.

364 FRANCES DEGEN HOROWITZ

States. The present state of affairs in the field can be traced to the late 1960s and early 1970s. At that time, by many accounts, S-R psychology and behaviorism and its strong environmental- ist orientation were, with the exception of the Skinnerian brand, eclipsed (overthrown?) by the organismic/cognitive revo- lution (Horowitz, 1987; Stevenson, 1983). At first this was fueled by American developmentalists' having discovered Pia- get. There followed an almost "gee whiz" response as it was shown that infants and young children were capable of much more complex behavior than had been previously supposed (Kessen, Haith, & Salapatek, 1970; Stone, Smith, & Murphy, 1973).

At the same time, the development of some new methodolo- gies involving techniques to study habituation in infants (Ber- lyne, 1958; Fantz, 1964) and assessments of neonatal behavior produced an explosion of information on infant capabilities. As a result of many demonstrations, it became obvious that there was a much more well developed behavioral repertoire in the newborn and young infant human organism than anyone had previously described (Brazelton, 1973; Kessen et al., 1970; Stone etal., 1973).

This growing body of evidence about the abilities of the in- fant and young child served to challenge the behaviorist as- sumption that learning accounted for the acquisition of early behavior. This evidence, coupled with the demonstration of the Piagetian phenomena, particularly in the realm of cognitive development through adolescence, called into question the en- tire behavioristic enterprise. If Gesell's name did not surface in the discussion of these matters, it was certainly a return, albeit in a more sophisticated framework, to some of Gesell's basic tenets.

Even as Piagetian theory was being modified, the discus- sions of systems theory and transactional theory applied to developmental theory offered attractive theoretical alternatives to behaviorism (Sameroff & Chandler, 1975; Sameroff, 1983). The infant was not a tabula rasa; experience could not write anything it wished. No one denied some role to learning, but it certainly was not considered a central mechanism. Basic re- search on learning in children declined precipitously. Stu- dies of children's learning were increasingly confined to re- search inspired by Skinnerian principles.

At the same time, the 1960s bloom of optimism about the power of early intervention programs to change developmental outcome began to fade (Clarke & Clark, 1976). Though many of these programs were not behavioristic in their programmatic orientation, the rationale for mounting them was strongly in- fluenced by the environmentalism that had predominated since the 1930s. However, the fledgling intervention efforts had failed to demonstrate dramatic changes in developmental out- come, school achievement, or IQ (Horowitz & Paden, 1973; Jensen, 1969). Advances in genetics identifying genetic contri- butions to behavior strengthened further the growing belief in organismically determined behavioral development and devel- opmental outcome. No one was suggesting that experience and the environment were irrelevant to behavioral development, but the pendulum had clearly swung away from environmental determinism and toward genetic and organismic determinism.

The inclination to cognitivism in developmental psychology and to organismic and genetic determinism is, however, just

that: an inclination rather than a reflection of consensus. In fact, clear theoretical labels are hard to come by among today's developmentalists. When the 45 respondents to the survey used in preparation for writing this essay were asked to identify how they classified themselves theoretically, a total of 26 different labels were used. The two most frequent classifications were eclectic and constructivist, each typifying four persons. Eight persons used some variation of cognitive: cognitive developmen- tal cognitive social learning, sociocognitive. Adding the one who used cognitive behaviorism, cognitive was by far the most fre- quently used theoretical term. Seven persons, however, used some form of behaviorist: descriptive behaviorist, behavior ana- lyst, liberalized S-R, ox social interactive behaviorist. One person proposed social evolutionary cognitive behaviorism as a pre- ferred label.

The respondents also did not agree about whether most current psychologists are, from a methodological point of view, functionally Watsonian behaviorists. A little over 60% dis- agreed or tended to disagree with such a claim. Thus, in this sample, there was a lack of agreement even about what many have felt was the methodological standard for the field. In re- sponse to a question asking whether the emphasis on learning and environment will reassert itself in the field, the respon- dents split almost evenly in their tending to agree or disagree that this was a likely possibility.

Although those responding to the questionnaire were not a random sample, almost all of the respondents were relatively senior developmentalists (see footnote 1), and they obviously represented a wide spectrum of opinion. The dispersion of self- describing theoretical labels and the ambivalence with respect to the likelihood of a return of emphasis on environment and learning suggest that developmental psychology is currently in a rather fluid theoretical period. In trying to identify Watson's legacy and the possibility of a Watsonian presence in the field in the coming years, one is struck by a number of contradictory possibilities.

Putting methodology aside, Watson's basic position was that the principles of learning would account for the largest share of behavioral development and developmental outcome and that these principles are exercised almost exclusively through envi- ronmental opportunities for children to learn. Though the theo- retical interest in children's learning has waned, in fact, there is a great deal of research on learning in the field of education as well as among those who identify themselves as Skinnerians. Though the belief that environment and experience are the main shapers of human behavioral development no longer ex- ists in its extreme form, there is, in fact, widespread acceptance of the idea that experience is important to development. Inter- vention programs continue to be mounted, albeit often without any recourse to an explicit use of learning principles even as the intervention appears to rest on the assumption that experience will make a difference. Environment and experience are given importance in recent discussions of cultural diversity and how cultural experience contributes to the shaping of the behavioral repertoire. However, the mechanisms by which this occurs are often not addressed (Horowitz, 1987).

Transactional theory (Sameroff & Chandler, 1975), dynami- cal systems theory (Thelen, 1990; Thelen, Kelso, & Fogel, 1987) as well as Vygotsky's theory (Brown, 1982; Rogoff, 1990) are

APA CENTENNIAL: JOHN B. WATSON'S LEGACY 365

heavily referenced these days and arrayed with the organismic approach. Yet an analysis of Vygotsky and of transactional theory and even of systems theory reveals that these ap- proaches give a healthy role to both the environment and to learning in an interaction with the organism. No developmen- talists totally exclude experience and the environment as vari- ables contributing to developmental outcome, so one wonders whether the negative reactions to Watson and behaviorism are focused still on Watson's persona or on what some have per- ceived as extreme Skinnerian claims for the power of the envi- ronment, or just on the simplistic nature of Watson's develop- mental approach.

One survey respondent stated that the misguided theoretical position of Watson and behaviorism was responsible for a loss of some 50 years of productive developmental research. Is it possible to ask whether the retreat from basic research on learn- ing in a developmental context, the dismissal of Watson and behaviorism, and the isolation of much of applied Skinnerian research has not had its costs?

Is the current quite fluid state of developmental theory re- lated to the lack of a clear vision about how to parse into the developmental equation a functional role for learning and envi- ronment in development? Is there now a sufficient body of new data that makes this a possibility?

There are a number of signs that the pendulum is swinging back to a middle position. The proposal for a structural/be- havioral model to account for development outcome (Horo- witz, 1987) suggests, among other things, a possible rapproche- ment between the behavioral and organismic emphases. It does this by focusing on the idea that a productive approach to un- derstanding behavioral development is to consider that there are two groups of behaviors that develop: those that are univer- sal behaviors and those that are nonuniversal behaviors.

In the structural/behavioral model, the universally acquired behaviors are defined as species typical. They have an evolu- tionary base, and they occur with almost 100% probability in all normal human organisms. Although some environmental transactions, perhaps learning, are necessary for their acquisi- tion, these behaviors are rooted in organismic characteristics. The nonuniversal behaviors are acquired only as learned behav- iors and are dependent on environmental opportunity, though they may have a base in the universal behavioral repertoire. In the adult repertoire the larger share of the repertoire probably is made up of nonuniversal behaviors. Furthermore, many of these define whether a person will be able to function produc- tively and successfully in a society or culture.

From the perspective of the structural/behavioral model a full understanding of development will not be achieved unless ways are found to account for the acquisition of both the univer- sal and nonuniversal behaviors. This means psychologists must include in their scientific agenda basic research on all the mech- anisms involved in learning.

Terms like contextualism, transaction, and transcontextual, as well as discussions of a developmental contextual model (Lerner, 1991), are broadening the perspective of developmen- tal psychology. Thus behavior and development are placed in a dynamic systems perspective in which the organism is in con- stant transaction with the environment, particularly with the social environment (Cairns, 1991). Environment, experience,

and the mechanisms that describe the organism-environmen- tal relationships must, ultimately, include some understanding of how learning operates in dynamic systems.

The strongest case currently being made for the role of experi- ence and environment in a systems context involves the recogni- tion of the mutually influential gene-environment relation- ships. Here it is acknowledged that genetic expression requires an environmental context and is affected by variations in envi- ronment (Oyama, 1985). Expanding on Waddington's (1966) notions of the canalization of development, Gottlieb (1991) has proposed that normally occurring experience can also serve the canalization process. These ideas bring us full circle back to Watson.

Watson believed that environmental shaping of behavior be- gan prenatally. His ideas stimulated Z-Y. Kuo, a skeptical Chi- nese scientist, to undertake in the 1930s and 1940s a series of experiments in which Kuo systematically altered the prenatal embryonic environments of chicks and other species to see if he could produce different behavioral repertoires. He found evi- dence for Watson's speculations. Due to the vagaries of life in China during those and subsequent years, Kuo's work did not become widely known in the United States until the 1960s and 1970s (Kuo, 1976).

Numerous studies supporting Kuo's data (and, indirectly, a number of Watson's ideas) have appeared. All of them demon- strate the same basic principle: Manipulations of prenatal and postnatal environments in a variety of animal species produce different patterns of behavior and affect behaviors once thought to be innate, genetically controlled, and unalterable (e.g., Gottlieb, 1978; Marler, 1977).

Watson's legacy to developmental psychology, aside from methodology, is the emphasis he placed on the importance of learning and experience in development and on the need to understand the principles by which learning and experience function. He insisted that learning and experience could be the sole elements determining development and took the position to its extremes. As soon as other data were seen as challenging Watson's assertions, dislike for both his dogmatism and his per- sona appears to have given critics broad license for dismissing and caricaturing him and all of behaviorism. His point of view has been labeled mechanistic, which is seen as synonymous with simplistic. YeX, mechanisms need not be simple, and com- plex systems are only understandable in terms of the mecha- nisms that account for their functioning. Is this likely to be less true for behavior?

What has flourished in the name of behaviorism has been largely associated with Skinner and with applied psychology, though not necessarily applied developmental psychology. Ap- plied behaviorism has been relatively isolated from the main- stream of developmental psychology. In the broadening view of contexts and transactions and dynamic interchange that ap- pears to be gaining in developmental psychology, will there be a return to some meaningful inclusion of the processes of learn- ing in the developmental research agenda?

Prognostication about the direction in which a scientific field will move is risky business. The strength of Watson and behaviorism in focusing on learning is that it is a focus on process. To understand behavior, development, contexts, and systems, we must ultimately understand the processes that ac-

366 FRANCES DEGEN HOROWITZ

count for the phenomena of interest. It is difficult to think that much progress will be made on understanding processes in development without understanding how the principles of learning operate across the life span.

Today's discussions of developmental theory and develop- mental psychology refer to an exceedingly more complex enter- prise than existed in Watson's time. We now have many more facts about both behavioral and biological events. The ability to relate behavioral and biological functioning and social context as interacting or transacting with one another and to think of doing this developmentally has put us on the threshold of a much more powerful developmental science than we might have envisioned even 10 years ago. If Watson has any enduring legacy to the field of developmental psychology beyond his methodological manifesto, it is currently to be found in his active descendants. Among these are a small group of experi- mental child psychologists working still with Hull-Spence theory (e.g., Cantor & Spiker, 1989) or actively studying learning in young children using experimental techniques (e.g., Rovee- Collier, 1986; Rovee-Collier, Earley, & Stafford, 1989). A larger group works within the Skinnerian tradition (e.g., Poulson, Nunes, & Warren, 1989; Riegler & Baer, 1989). Watson's influ- ence may continue in the future among those who elect to bring back into mainstream developmental psychology an emphasis on understanding the principles of learning and how they oper- ate in developmental processes.

Watson's methodological position has been widely regarded as his most enduring influence. Yet, many psychologists today deny that they are methodological behaviorists. It is not clear whether this is a rejection of the label or of the methodological tenets themselves. A scan of the basic psychological journals reveals, still, a general use of the standards of independently verifiable observations in ways obviously Watsonian, though not necessarily involving only experimental paradigms. Wat- son's point of view on experience and environment has been moderated by most contemporary psychologists and some- times dismissed. Yet there is considerable evidence that the influence of experience and the environment on development are acknowledged even as the mechanisms by which they oper- ate are not articulated or a focus of much theoretically sophisti- cated research. Watson's S-R psychology is clearly of the past. Yet, even in complex systems there are relationships of stimuli and responses for which psychologists need to account.

John Broadus Watson was a controversial man. His theory was controversial. His advocacy for an uncompromising posi- tion on the role of the environment was extreme. It will be to the benefit of developmental psychology if we can finally over- come his persona, his ill-advised extremism, and the unpro- ductive caricaturing of both in favor of a greater understanding and appreciation of the roles that learning and experience play in behavioral development. The data, if not human tolerance, ought to push us in that direction.

References

Bergmann, G. (1956). The contribution of John B. Watson. Psychologi- cal Review, 63, Ite-llb.

Berlyne, D. E. (1958). The influence of the albedo and complexity of

stimuli on the visual fixation in the human infant. British Journal of Psychology, -#9, 315-318.

Bijou, S. W (1955). A systematic approach to an experimental analysis of young children. Child Development, 26,161-169.

Bijou, S. W (1957). Patterns of reinforcement and resistance to extinc- tion in young children. Child Development, 28, 47-54.

Bijou, S. W (1958). Operant extinction after fixed interval schedules with young children. Journal of Experimental Analysis of Behavior, 1, 25-49.

Bijou, S. W, & Baer, D. M. (1961). Child development (Vol. 1). New York: Appleton-Century-Crofts.

Bijou, S. W, & Baer, D. M. (1963). Some methodological contributions from a functional analysis of child development. In L. P. Lipsitt & C. C. Spiker (Eds.), Advances in child development andbehavior (Vol. 1, pp. 147-196). San Diego, CA: Academic Press.

Brazelton, T. B. (1973). Neonatal Behavioral Assessment Scale. Philadel- phia: Lippincott.

Brown, A. L. (1982). Learning and development: The problems of com- patibility, access and induction. Human Development, 25, 89-115.

Buckley, K. W (1989). Mechanical man: John Broadus Watson and the beginnings of behaviorism. New York: Guilford Press.

Cairns, R. B. (1991). Multiple metaphors for a singular idea. Develop- mental Psychology, 27, 23-26.

Cantor, J. H. & Spiker, C. C. (1989). Children's learning revisited: The contemporary scope of modified Spence discrimination theory. In H. W Reese (Ed.), Advances in child behavior and development(Vol. 21,121 -151). San Diego, CA: Academic Press.

Clarke, A. M., & Clarke, A. B. D. (1976). Early experience: Myth & evidence. New York: Free Press.

Dollard, J., & Miller, N. E. (1950). Personality and psychotherapy. New York: McGraw-Hill.

Fantz, R. L. (1964). Visual experience in infants: Decreased attention to familiar patterns relative to novel ones. Science, 146, 688-690.

Gottlieb, G. (1978). Development of species identification in duck- lings. Change in species-specific perception caused by auditory de- privation. Journal of Comparative & Physiological Psychology, 92, 375-387.

Gottlieb, G. (1991). Experiential canalization of behavioral develop- ment: Theory. Developmental Psychology, 27, 4-13.

Horowitz, F. D. (1987). Exploring developmental theories: Toward a structural/behavioral model of development. Hillsdale, NJ: Erlbaum.

Horowitz, F. D, &Paden, L. Y. (1973). The effectiveness of environmen- tal intervention programs. In B. M. Caldwell & H. N. Ricciuti (Eds.), Review of child development research (Vol. 3, pp. 331-401). Chicago: University of Chicago Press.

Hull, C. L. (1943). Principles of behavior. New York: Appleton-Century- Crofts.

Jensen, A. R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 39,1-123.

Jones, M. C. (1924a). The elimination of children's fears. Journal of Experimental Psychology, 7, 382-390.

Jones, M. C. (1924b). A laboratory study of fear: The case of Peter. Pedagogical Seminary, 31, 308-315.

Kendler, H. H., & Kendler, T. S. (1961). Effect of verbalization on discrimination rehearsal shifts in children. Science, i j ^ , 1619—1620.

Kendler, H. H., & Kendler, T. S. (1969). Reversal-shift behavior: Some basic issues. Psychological Bulletin, 72, 229-232.

Kendler, T. S., & Kendler, H. H. (1959). Reversal and nonreversal shifts among kindergarten children. Journal of Experimental Psychology, 58, 56-60.

Kendler, T. S., & Kendler, H. H. (1966). Optional shifts of children as a function of number of training trials on the initial discrimination. Journal of Experimental Child Psychology, 3, 216-224.

Kendler, T. S., Kendler, H. H., & Learnard, B. (1962). Mediated re-

APA CENTENNIAL: JOHN B. WATSON'S LEGACY 367

sponses to size and brightness as a function of age. American Journal of Psychology, 75, 571-586.

Kessen, W, Haith, M., & Salapatek, P. (1970). Human infancy: A bibli- ography and guide. In P. H. Mussen (Ed.), Carmichael's manual of child psychology (Vol. 1, 3rd ed., pp. 287^47). New York: Wiley.

Krasnogorski, N. (1909). Ueber die Bedingungsreflexe im Kindesalter [Conditioned reflexes in childhood]. Jahrbuch fur Kinderheilkunde und Physische Erziehung, 19,1-24.

Kuenne, M. R. (1946). Experimental investigation of the relation of language to transposition behavior in young children. Journal of Experimental Psychology, 36, 471-490.

Kuo, Z-Y. (1976). The dynamics of development (Rev. ed.). New York: Plenum Press.

Lerner, R. M. (1991). Changing organism-context relations as the ba- sic process of development: A developmental contextual perspec- tive. Developmental Psychology, 27, 27-32.

Mateer, F. (1918). Child behavior: A critical and experimental study of young children by the method of conditioned reflexes. Boston: Badger.

Marler, P. (1977). Development and learning of recognition systems. In T. H. Bullock (Ed.), Recognition of complex acoustic signals. Berlin, Germany: Dahlem Konferenzen.

McCandless, B. R., & Spiker, C. C. (1956). Experimental research in child psychology. Child Development, 27, 75-80.

Miller, N. E. (1941). The frustration-aggression hypothesis. Psychologi- cal Review, 45,337-342.

Miller, N. E., & Dollard, J. (1941). Social learning and imitation. New Haven, CT: Yale University Press.

Nance, R. D. (1970). G. Stanley Hall & John B. Watson as child psychol- ogists. Journal of the History of the Behavioral Sciences, 6, 303-316.

Oyama, S. (1985). The ontogeny of information. Cambridge, England: Cambridge University Press.

Poulson, C. L., Nunes, L. R. de Paula, & Warren, S. F. (1989). Imitation in infancy: A critical review. In H. W Reese (Ed.), Advances in child development and behavior (Vol. 21, pp. 271-298). San Diego, CA: Academic Press.

Riegler, H. D., & Baer, D. M. (1989). A developmental analysis of rule- following. In H. W Reese (Ed.), Advances in child development and behavior (Vol. 21, pp. 191-219). San Diego, CA: Academic Press.

Rogoff, B. (1990). Apprenticeship in thinking. Oxford, England: Oxford University Press.

Rovee-Collier, C. (1986). The rise and fall of infant classical condition- ing research: Its promise for the study of early development. In L. P. Lipsitt and C. Rovee-Collier (Eds.), Advances in infancy research (Vol. 4, pp. 139-159). Norwood, NJ: Ablex.

Rovee-Collier, C, Earley, L., & Stafford, S. (1989). Ontogeny of early event memory: III. Attentional determinants of retrieval at 2 and 3 months. Infant Behavior and Development, 12,147-162.

Sameroff, A. J. (1983). Developmental systems: Contexts and evolu- tion. In P. H. Mussen (Ed.) & W Kessen (Vol. ed.), Handbook of child psychology: Vol. 1. History, theory and methods (pp. 273-294). New York: Wiley.

Sameroff, A. X, & Chandler, M. (1975). Reproductive risk and the con- tinuum of caretaking casualty. In F. D. Horowitz (Ed.), Review of child development research (Vol. 4, pp. 187-244). Chicago: University of Chicago Press.

Sears, R. R. (1951). A theoretical framework for personality and social behavior. American Psychologist, 6, 476-483.

Skinner, B. F. (1938). The behavior of organisms: An experimental analy- sis. New York: Appleton-Century-Crofts.

Skinner, B. F. (1948). Walden two. New York: Macmillan. Skinner, B. F. (1953). Science and human behavior. New York: Macmil-

lan. Skinner, B. F. (1959). John Broadus Watson, behaviorist. Science, 129,

197-198. Skinner, B. F. (1969). Behaviorism at fifty. In B. F. Skinner (Ed.), Con-

tingencies of reinforcement: A theoretical analysis (pp. 221 -268). New York: Appleton-Century-Crofts.

Skinner, B. F. (1974). About behaviorism. New York: Knopf. Stevenson, H. (1983). How children learn—The quest for a theory. In

P. H. Mussen (Ed.) & W Kessen (Vol. ed.), Handbook of child psychol- ogy: Vol. 1. History, theory and methods (pp. 213-236). New York: Wiley.

Stone, X, Smith, H., & Murphy, L. (Eds.). (1973). The competent infant. New York: Basic Books.

Thelen, E. (1990). Dynamical systems and the generation on individual differences. In J. Colombo & J. Fagen (Eds.), Individual differences in infancy: Reliability, stability, prediction (pp. 19-43). Hillsdale, NJ: Erlbaum.

Thelen, E., Kelso, J. A. S., & Fogel, A. (1987). Self-organizing systems and infant motor development. Developmental Review, 7, 39-65.

Waddington, C. H. (1966). Principles of development and differentia- tion. New York: Macmillan.

Watson, J. B. (1913). Psychology as the behaviorist views it. Psychologi- cal Review, 20,158-177.

Watson, X B. (1919). Psychology from the standpoint of a behaviorist. Philadelphia: Lippincott.

Watson, X B. (1970). Behaviorism (Rev. ed.). New York: Norton. (Origi- nal work published 1924)

Watson, X B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 5,1-14.

Received November 25,1991 Revision received January 2,1992

Accepted January 3,1992 •