Introduction

The study of development poses a major paradox for students of behavior. The problem arises from the fact that change is an essential property of development. Virtually all features of the organism undergo modification during its life span. On the other hand, continuity over time seems essential for individual uniqueness, organization, and the maintenance of integrated patterns of behavior. The paradox is simply this: How can continuity and persistence be achieved in an organismic system that necessarily undergoes maturational, interactional, and cultural-social change?

“This question brings us directly to the problem of continuity and discontinuity in development. Resolution of this problem will depend on a number of issues. These include clear definitions of exactly what is meant by a developmental continuity or discontinuity, identification of a research strategy that can yield results that arrive at conclusions about continuity consistent with these definitions, analysis of the types of measures and inferential methods that can be used to arrive at such conclusions, and examination of the longitudinal research designs that are required for studying developmental continuities.”Footnote ¹

When I began my research career in developmental psychology a half-century ago, the field seemed to be facing several daunting challenges even beyond those embodied in the above-described paradox. Within the field there was clearly a lack of consensus regarding basic theoretical frameworks, substantial disagreements about fundamental developmental concepts, and serious concerns about methodological rigor and sophistication. No single overarching theory dominated the research enterprise; instead, adherents of competing schools of thought, for example, social learning versus psychoanalysis, seemed to operate in parallel universes, each with its own set of basic concepts and assumptions regarding developmental phenomena. Arguments about the origins of both developmental phenomena and developmental processes, that is, nature versus nurture issues, permeated much of the field. Additional controversies centered on the nature, prevalence, and persistence of critical periods, as well as the relative distinctiveness of different developmental phases, as in Piagetian stages.

At the same time, efforts to assess the relative stability of behavioral patterns, cognitive capabilities, and biological processes across developmental time likewise faced formidable challenges. Regarding issues of basic experimental design and measurement, there were relatively few longitudinal studies across the entire field, and most were retrospective rather than prospective. Consequently, the bulk of information regarding developmental change largely came from cross-sectional comparisons and personal memories, thereby rendering most hypotheses regarding possible mechanisms underlying observed age differences little more than informed speculation, no matter what measures were being employed. Additionally, nature-nurture arguments about measurement remained as vigorous and uncompromising as they had been a century earlier in the day of Darwin’s Origin of the Species, impacting multiple research topics ranging from the relative heritability of IQ to the perceived malleability of gender identity. All in all, the field of developmental psychology seemed ripe for change back in the early 1970s.

Developmental psychology was not the only scientific domain seemingly in a state of flux during that period. Across the biological sciences, the relatively new field of ethology was vividly spotlighted by the 1973 Nobel Prize awarded to Lorenz, Tinbergen, and von Friest for their studies of animal social behavior; two years later the field of evolutionary biology was shaken up by E. O. Wilson’s Sociobiology, ironically ginning up heated arguments about the origins of various aspects of human behavior based largely on studies of ants. Contemporaneously, the child psychiatry world was likewise undergoing a major revolution. Traditional psychoanalytic approaches, dominant for decades, were quite suddenly almost totally eclipsed by more biologically oriented approaches to theory, research, and practice, soon blossoming into specialties oriented around neuroscience, psychopharmacology, neurophysiology, and ethology – and out of ethology came a new emphasis on the concept of attachment, a burgeoning movement that emerged from John Bowlby’s ideas about mother-infant relationships. It didn’t take long for that movement to turn into a virtual stampede, eventually accepted and even adopted by a major segment of the developmental psychology community, and it has remained highly influential to this day.

It has been my great pleasure and honor to have participated in this movement throughout my career, starting in the very laboratory where a decade earlier Harry Harlow’s pioneering research with rhesus monkeys had famously first provided powerful empirical support for Bowlby’s original ideas. The research journey that my colleagues and I pursued initially expanded Harlow’s focus on attachment-related phenomena in monkeys to include a wider range of social activities and, whenever feasible, to characterize associated biological activity, thereby generating a host of novel findings and concepts that both broadened relevant perspectives and opened new pathways for investigation. This chapter describes some of the highlights of that journey, with special emphasis on the unique sets of colleagues and circumstances that served to generate major inflection points in shaping the scope and direction of that research. The result has been a body of work that has undergone several dramatic changes over time yet retained its overall focus on the development of social capabilities and concomitant biological processes in primate species, including humans, thereby maintaining a lasting continuity throughout.

Early Years at Wisconsin: Bowlby, Hinde, and Harlow

I began my research career in developmental psychology by reading the correspondence between John Bowlby, also known as “the father of attachment theory,” and Harry F. Harlow, also known as “the father of the cloth monkey mother.”

Bowlby was a psychoanalytically oriented psychiatrist by training, well-known by pedigree (his father had served as Physician to the King) but he was, in fact, a passionately dedicated ethologist at heart, as evidenced by his richly detailed and insightful clinical observations of behavior in infants and young children. Given his strong biological background and an ever-keen interest in ethological studies of developmental phenomena in animals, Bowlby actively sought out support for his ideas largely through extended discussions with his close friend and colleague Robert A. Hinde. Hinde, a noted Cambridge biologist who had begun studying mother-infant interactions in rhesus monkeys at that time, enthusiastically promoted Bowlby’s emerging ideas about attachment, an endorsement that carried considerable academic clout among his fellow ethologists.

In the mid 1950s Hinde formally introduced Bowlby to Harry Harlow at a scientific meeting in London, and Harlow and Bowlby soon thereafter began active correspondence, keeping each other abreast of their latest findings and ideas, meeting up at professional gatherings, and even visiting each other’s facilities. Years later, as a Harlow graduate student at Wisconsin, I was given the distinct privilege of reading some of this correspondence, along with selected (p)reprints containing handwritten notes and comments scribbled throughout. Needless to say, some of this personal archival material left deep impressions that have remained throughout my own professional career.

That career didn’t exactly start from scratch. When I finished my degree at the University of Wisconsin–Madison Harlow told me to “stick around because I’ll be retiring soon.” I told him I would; and three years later he did, but not before essentially giving me the keys to the lab he had founded thirty plus years earlier, albeit with some administrative legerdemain and influential proactive support from Robert Hinde. The following year I joined the same academic department (Psychology) from which Harlow had just retired.

Harlow’s lab had been arguably one of the world’s best places to study primate behavior for decades, with a research and administrative staff second to none in terms of experience, expertise, and dedication, including Steven Eisele, who ran the breeding colony, Christopher Ripp, who oversaw the neonatal nursery operation, and Helen LeRoy, who had become chief administrator after serving as Harlow’s personal secretary for years. The lab was frequently visited by numerous prominent scientists, including former Harlow PhDs, most notably (for me) Leonard A. Rosenblum, several world-class developmental psychologists, including Michael Lewis and William T. Hartup, and famous scientists from a variety of other fields, including the Robert Hinde. Harlow also strongly encouraged me to present papers at scientific meetings, especially those focusing on human child development, and he redirected several of his numerous speaking invitations from outside organizations to me instead. In retrospect, Harlow’s postgraduate mentoring was probably as important for my professional development as anything I had learned in graduate school.

In the years immediately following Harlow’s retirement, our research team continued follow-up studies of interventions for socially deprived infant monkeys (e.g., Suomi, Harlow, & Novak, Reference Suomi, Harlow and Novak1974), explored other types of social relationships, especially those involving peers, and in collaboration with William T. McKinney’s research group in the UW Department of Psychiatry, began preliminary investigations of biological correlates of various behavioral patterns, culminating in long-term studies primarily directed by Gary W. Kraemer. Little did I realize at the time how unusual such a cross-disciplinary collaboration really was. Crucially, we also began redirecting much of our research efforts to studies of infant and juvenile monkeys reared in more socially enriched settings, such as the ingenuous “nuclear family” set-up created by Margaret K. (Peggy) Harlow before she passed in 1971. This shift in emphasis, one designed to better understand species-normative social development, represented a major turning point for our overall research program at the time.

The Bielefeld Project and Beyond

In the fall of 1977, the Center for Interdisciplinary Research (ZIF) at the University of Bielefeld in (West) Germany began an ambitious nine-month project focusing on early behavioral development in humans and (other) animals. The overall project was headed by Klaus Immelmann, Chair of Ethology at the University of Bielefeld, internationally acclaimed for his pioneering research on sexual imprinting in Zebra finch. Immelmann was given sufficient resources to bring together a remarkable collection of senior researchers from multiple disciplines, literally a who’s who in each discipline, all sharing a common interest in developmental phenomena. In addition to ethology and developmental psychology, disciplines represented included evolutionary biology, cognitive science, physical and cultural anthropology, developmental neurobiology, and child and adolescent psychiatry. Some participants remained on site for the entire project; others came for shorter periods, sometimes repeatedly. The agenda included extended seminars presented by these senior scientists, as well as extensive discussions following each presentation and later within ad hoc subgroups. It was akin to receiving an almost-daily tutorial from some of the world’s very best in the business.

The official end-product of the project was an edited volume, largely consisting of co-authored chapters from these various subgroups (Immelmann et al., 1981). Ross Parke and I co-authored one chapter surveying paternal behavior in primates, including humans. In another chapter, Immelmann and I examined so-called critical periods and imprinting-like phenomena, and we ended up proposing a general synthesis in which classic avian filial and sexual imprinting, with their sharply defined critical periods and apparent permanence, merely represented extreme cases within a more general concept of sensitive periods that had widespread prevalence well beyond that of classic imprinting paradigms (Immelmann & Suomi, 1981). In the process we developed a close professional relationship and personal friendship that lasted a lifetime.

Finally, there was a chapter co-authored by G. P. (Jim) Sackett, Arnold Sameroff, Robert Cairns, and myself, the opening paragraphs of which are presented at the start of this one. Much of the chapter involved the creation of a hypothetically ideal experimental design as a template for rigorous longitudinal investigation of virtually any type of developmental phenomena involving virtually any set of data collection and measurement techniques. Although the immediate focus was on developing a means of operationalizing Sameroff’s concept of transactions among both persons and environments, the template seemed applicable to many other circumstances as well. It represented a longitudinally oriented methodologist’s ultimate dream developmental design. None of us really thought then that our idealized model design would ever come to fruition; it was at best merely a “thought experiment.”

My on-site time at the ZIF totaled perhaps two months, but its impact was substantial and lasting, generating other turning points in the research agenda. Back in Madison, the research team began expanding and extending longitudinal collection of behavioral data from infancy onward in monkeys with different early social attachment opportunities; and we found long-term relationships with multiple behavioral indices of social competence later in development. Importantly, this emerging longitudinal data base also facilitated discovery of developmentally stable individual differences in measures of temperament independent of social rearing background, and we discovered those differences were predictive of later differences in social competency as development progressed. Moreover, the patterns of individual differences exhibited by some of our infant monkeys resembled those of human infants and toddlers characterized by Jerome Kagan and others as being “behaviorally inhibited.”

The Bielefeld experience produced another turning point in the lab’s research focus, a push to study developmental phenomena in settings made to be as naturalistic as possible. We began by moving a small group of juvenile monkeys into a large corncrib situated on a farm outside of Madison, basically at the behest of Peggy O’Neil (Wagner), a graduate student residing on the farm, who had studied those monkeys back in the lab from birth onward. She soon surreptitiously discovered that the monkeys could be allowed to exit their corncrib and free-range over the rest of the farm as they pleased as long as the corncrib remained accessible as a “secure base” to which they could return (or be returned) upon demand. This chance discovery unexpectedly opened a wide range of research opportunities that could be conducted in a field-like setting, even one in rural Wisconsin!

Yet, conducting such research, as well as expanding the scope of the longitudinal research projects then underway back inside the main Primate Lab building, continued to present daunting challenges, even for an unusually well-equipped, superbly staffed, and generously funded research facility. Underlying many of the challenges was a continual need for long-range external support to maintain the longitudinal studies and access to other resources that UW seemingly was unable to provide at that time. It was then, in early 1983, that I was first contacted by Arthur S. Levine, the recently appointed Scientific Director at NICHD.Footnote ² Over the next few months he put together an offer that I simply could not refuse, one that turned out to be probably the most consequential turning point of my professional career.

Art Levine, the LCE, and Poolesville Lab

When Art Levine began his tenure at NICHD he decided to support the creation of a new intramural laboratory that could conduct translational research to help bridge gaps between the basic research laboratories and clinically oriented branches of the Institute. Levine believed that one promising approach would be to create a major program centered around the development of nonhuman primate models. He named his proposed new lab the Laboratory of Comparative Ethology (LCE), he identified a location where such a lab could be developed on the grounds of the NIH Animal Center, near the town of Poolesville about thirty miles from the main NIH Bethesda campus,Footnote ³ where space available for facility development and renovation seemed limitless, and he recruited me to first develop and then head the new laboratory. Levine backed up his promises with real, tangible resources, even as he garnered additional resources from other NIH Institutes, utilizing creative administrative strategies to leverage for the unique laboratory space, equipment, and personnel requirements of the new lab. The end result was a Poolesville facility that supported a self-sufficient breeding colony, one that included a research-friendly neonatal nursery, a five-acre outdoor field enclosure, and indoor-outdoor pens for social group housing of the rest of the LCE rhesus monkey colony.

The cross-institute collaboration Levine promoted is best exemplified by the one he arranged with Markku Linoila, a new laboratory chief at NIAAAFootnote ⁴ who, like me, was setting up his own lab. Linoila was interested in monkeys who differed in observed aggressiveness, and I was interested in the CSF monoamine metabolite concentrationsFootnote ⁵ those monkeys might carry. So we quickly agreed to develop projects for which he would collect and analyze repeated CSF samples from my monkeys of interest. To “seal the deal” Markku offered tenure-track support for a junior candidate of my choice whose sole research role would be to oversee and manage this and subsequent collaborative projects. I chose J. D. Higley, my senior graduate student at UW, and he agreed to the arrangement. Two other UW graduate students, Maribeth Champoux and Gayle Byrne, as well as Peggy Wagner, also made the move to Poolesville, each subsequently establishing a long-term career at the NIH. For the next thirty-five plus years Dee Higley maintained his collaborative relationship with my laboratory, even after retiring from NIAAA after twenty years to assume a senior professorship at Brigham Young University. His senior postdoctoral fellow at the time, Allyson J. Bennett, was then promoted into his position as chief liaison with the LCE, followed several years later by Christina S. Barr when Allyson herself moved, first to Wake Forest and then to UW-Madison, where she was hired into Harlow’s old position in Psychology.

In the early days at Poolesville Higley helped coordinate much of the monkey move from Madison, and he subsequently directed the arrangement of social groups and rearing assignments of infants born into the colony, all the while carrying out his own program of independent research. Within a few years these housing and infant and social group assignment practices and procedures had achieved a critical mass of well-matched groups of differentially reared monkeys that could now be studied longitudinally from birth, both behaviorally and biologically. In other words, we had established the basic template needed to actually carry out those very kinds of prospective longitudinal studies envisioned by Sackett et al. (Reference Sackett, Sameroff, Cairns, Suomi, Immelmann, Barlow, Petrinovich and Main1981) in their dream developmental research design!

We have since exploited these capabilities for over three decades by encouraging a wide range of well-established researchers to join our ongoing longitudinal studies in order to collect their own favored measures of longitudinal change. These collaborative arrangements have led to hundreds of co-authored publications over the years. The master database continues to generate material for additional original publications to this day.

A prime example of using this body of data to generate entire programs of research can be seen in our collaboration with Peter Lesch and his colleagues based in Würzburg investigating specific gene-environment interplay (G x E) in monkeys. Lesch was the first to characterize a specific polymorphism in the human 5-HTTLPR geneFootnote ⁶ once thought to be associated with risk for depression; he subsequently identified a seemingly homologous polymorphism in our rhesus monkey colony.

We initially found significant interactions between early rearing conditions and allele type (G x E interactions) for measures of serotonin metabolism (Bennet et al., Reference Bennett, Lesch, Heils, Long, Lorenz, Shoaf, Champoux, Suomi, Linnoila and Higley2002), as well as measures of neonatal activity (Champoux et al., Reference Champoux, Bennett, Shannon, Higley, Lesch and Suomi2002). We then expanded the range of outcomes for which significant G x E interactions involving the 5-HTTLPR polymorphism and differences in early rearing conditions appear, including social play, and behavioral reactions to a variety of social stressors, as well as in epigenetic regulation of brain activity. In addition, we reported significant G x E interactions between early rearing and polymorphisms for several other candidate genes. In virtually every case a similar pattern was observed: the less efficient (transcription-wise) allele was associated with a negative outcome among nursery-reared monkeys, but a neutral or, in some cases, even an optimal outcome, for their maternally reared counterparts carrying that same less efficient allele. This consistent pattern suggests an overall (positive) buffering effect of competent early parenting for individuals carrying these so-called risk alleles. These findings, and others, give us greater understanding than we might have ever thought possible of the interactions between biology/genetics and environment/experience in shaping the course of human development (= epigenesis).

All the studies of G x E were entirely dependent on the existence of our master data set, representing only one of the many collaborations with other research groups, both inside and outside the NIH, that have turned out to be highly productive over the past two decades. These include an NIH-based series neuroimaging studies of brain structure and function; assessments of chronic HPAFootnote ⁷ activity in different physical and social environmental settings in long-term collaborations with Melinda Novak, Jerold Meyer, and Amanda Dettmer; and investigations of epigenetic changes throughout development among differentially reared and genetically characterized monkeys, in collaboration with Moshe Szyf’s lab at McGill and the Chicago group led by Jim Heckman and Steven Cole. Additional collaborative programs focused on systematic investigation of mirroring activity in rhesus monkey neonates, while other collaborations have examined relative adaptive capabilities in different species of primates.

Developmental Research Network Participation

The burgeoning expansion of LCE collaborations produced an increasing demand for direct dissemination of these new findings to facilitate their integration to existing bodies of knowledge. One solution was to join different developmentally oriented research networks that were first emerging in the late 1980s. I became an active participant in several, including one in the Canadian Institute for Advanced Research, the brainchild of Sir Fraser Mustard, as well as a sequence of NIMH-sponsoredFootnote ⁸ groups seeking to understand “how experience gets under the skin,” for which Megan Gunnar was a driving force, as well as two networks exploring modelling approaches and strategies for moderating aging processes in individuals who experienced early social adversity. The most successful networks all had a core group of eight to twelve senior participants who would meet two or three times over at least a three-year period, renewable when deemed appropriate. Each was composed of scientists who represented multiple discipline but who had also chosen to focus on specific topics of common interest. No specific goal or end-product beyond expansion of a knowledge base seemed to be required for a network to be successful. Instead, relative success seemed to be a product of the animated discussions that emerged as participants got to know each other better. In my mind these networks represented one of the most efficient, if not the most enjoyable, means of expanding our basic knowledge base in these areas of interest. In the best case, institutional support for these types of networks will continue in future years.

Epilogue

It has been a full fifty years since I embarked on my professional career as a basic researcher in developmental psychology, and five years since I officially retired from NICHD. So much has changed over that research lifetime, especially in our basic understanding of developmental phenomena and our ability to identify and characterize the many changes found in almost every area under investigation.

Our research technological capabilities have been advanced far beyond the wildest dreams of even our most enthusiastic and innovative research pioneers from previous generations of developmental researchers, especially in terms of our ability to witness how experience actually “gets under the skin,” and to document both the short- and long-term consequences of that experience doing so. Our advances in data collection, analyses, and mathematical modelling certainly open us to new worlds of numbers and symbols but also carry the risk of overwhelming even the most brilliant mathematically oriented minds with extreme data overload.

The many lessons I have learned throughout my research career include a great appreciation of the incredible added value that well-regulated interdisciplinary efforts can bring to many areas of research, as well as the importance of creating circumstances that foster the development of long-lasting friendships among those collaborating scientists. Finally, many of the issues and controversies that dominated the field fifty years ago are still with us, albeit in more nuanced form: witness the continued arguments about nature versus nurture issues, even though we are now able to identify specific genes and characterize various environments, capabilities that were hardly anywhere on anybody’s radar back then. So clearly there has been both continuity and change throughout our field over the past half century. One can only wonder if that same paradox will still hold true fifty years from now as well.