David Baldwin's Trauma Information Pages

The Effects of a Secure Attachment Relationship on Right Brain Development, Affect Regulation, & Infant Mental Health

Abstract

Over the last ten years the basic knowledge of brain structure and function has vastly expanded, and its incorporation into the developmental sciences is now allowing for more complex and heuristic models of human infancy. In a continuation of this effort, in this two part work I integrate current interdisciplinary data from attachment studies on dyadic affective communications, neuroscience on the early developing right brain, psychophysiology on stress systems, and psychiatry on psychopathogenesis in order to provide a deeper understanding of the psychoneurobiological mechanisms that underlie infant mental health.

In this paper I detail the neurobiology of a secure attachment, an exemplar of adaptive infant mental health, and focus upon the primary caregiveris psychobiological regulation of the infantis maturing limbic system, the brain areas specialized for adapting to a rapidly changing environment. The infantis early developing right hemisphere has deep connections into the limbic and autonomic nervous systems and is dominant for the human stress response, and in this manner the attachment relationship facilitates the expansion of the childis coping capacities. This model suggests that adaptive infant mental health can be fundamentally defined as the earliest expression of flexible strategies for coping with the novelty and stress that is inherent in human interactions. This efficient right brain function is a resilience factor for optimal development over the later stages of the life cycle.

The fundamental importance of the psychological as well as the biological health of the infant has long been held as a cardinal principle by every clinical discipline that deals with young children - infant psychiatry, behavioral pediatrics, child psychology, developmental psychoanalysis, and more recently the emerging fields of developmental psychopathology and infant mental health. And yet a more precise characterization of the concept of infant mental health, like the definition of "mental health" itself, has been elusive. Theoretically, it is clear that there must be links between infant and adult mental health, yet these too have been ill-defined. Although there is a large body of clinical knowledge in psychiatry, abnormal psychology, and psychoanalysis affirming the centrality of early relational experiences on enduring adaptive and maladaptive aspects of personality, there has been some question as to the structural mechanisms by which such events positively or negatively influence the process of development as it continues over the life span. In other words, how do the earliest interactions between a maturing biological organism and the social environment influence infant mental health, what are the central functions that define infant mental health, and how does it influence mental health at later stages of development?

The defined mission of The Infant Mental Health Journal is to focus upon infant social-emotional development, caregiver-infant interactions, contextual and cultural influences on infant and family development, and all conditions that place infants and/or their families at risk for less than optimal development. In this work I want to suggest that although the unique importance of "optimal development" has long been addressed by the psychological sciences, due to the advances of "the decade of the brain," developmental neuroscience is now in a position to offer more detailed and integrated psychoneurobiological models of normal and abnormal development. The incorporation of this information into developmental psychological models could forge closer links between optimal brain development and adaptive infant mental health, as well as altered brain development and maladaptive mental health.

A theoretical concept that is shared by an array of basic and clinical sciences is the concept of regulation (Schore, 1994; 1996; 1998d; 1999c; 2000b), and because it integrates both the biological and psychological realms, it can also be used to further models of normal and abnormal structure-function development, and therefore adaptive and maladaptive infant mental health. Interdisciplinary research and clinical data are affirming the concept that in infancy and beyond, the regulation of affect is a central organizing principle of human development and motivation. In the neuroscience literature Damasio asserts that emotions are the highest order direct expression of bioregulation in complex organisms (1998), and that primordial representations of body states are the building blocks and scaffolding of development (1994). Brothers argues that emotion occurs "in the context of evolved systems for the mutual regulation of behavior, often involving bodily changes that act as signals" (1997, p. 123). Emotions and their regulation are thus essential to the adaptive function of the brain, which is described by Damasio:

The overall function of the brain is to be well informed about what goes on in the rest of the body, the body proper; about what goes on in itself; and about the environment surrounding the organism, so that suitable survivable accommodations can be achieved between the organism and the environment (1994, p. 90).

In a number of works I have described the earliest ontogeny of these adaptive brain functions, and have argued that the essential events that allow for the emergence of the regulatory systems that control such functions occur during the brain growth spurt (Schore, 1994; 1996; 1997b, 1998a, b, 2000b, d). Moreover, I have offered data which suggests that the inceptive stages of development represent a maturational period of specifically the early maturing right brain, which dominant in the first three years of human life (Schore, 1994; Chiron et al., 1997). The right brain is centrally involved in not only processing social-emotional information, facilitating attachment functions, and regulating bodily and affective states (Schore, 1994, 1998a), but also in the control of vital functions supporting survival and enabling the organism to cope actively and passively with stress (Wittling & Schweiger, 1993). Furthermore, in a series of contributions I have proposed that the maturation of these adaptive right brain regulatory capacities is experience-dependent, and that this experience is embedded in the attachment relationship between the infant and primary caregiver (Schore, 1994; 1999b; 2000a, b; in press, c ). But it is important to point out that this experience can either positively or negatively influence the maturation of brain structure, and therefore the psychological development of the infant. This developmental psychoneurobiological model clearly suggests direct links between secure attachment, development of efficient right brain regulatory functions, and adaptive infant mental health, as well as between traumatic attachment, inefficient right brain regulatory function, and maladaptive infant mental health.

In an attempt to forge these conceptual links more tightly, in this two-part work I will address the problem of operationally defining adaptive and maladaptive infant mental health by integrating very recent data from attachment theory, developmental neuroscience, and developmental psychopathology. The primary goal of this latter field is to characterize the ontological processes whereby early patterns of individual adaptation evolve into later patterns of adaptation (Cicchetti, 1994), and thereby it investigates the early development of the individualis coping systems. In generating models of how early ontogenetic factors predispose high-risk individuals to later psychopathologies, this rapidly growing interdisciplinary approach is directly inquiring into the mechanisms that account for the continuity between infant mental health and mental health at later points in the life span.

An essential principle of the developmental psychopathology perspective is that atypical development can only be understood in the context of typical development, and so the focus is on underlying mechanisms common to both. This model suggests that any overarching conception of early development needs to integrate both the biological and psychological realms, and that it must incorporate models of both adaptive and maladaptive infant mental health. It also implies that infant mental health can not be defined solely as a "psychological" construct - rather, it is more precisely characterized as "psychobiological." Utilizing such a perspective, in these two papers I will contrast the neurobiology of a secure attachment, an exemplar of adaptive infant mental, with the neurobiology of an insecure disorganized/disoriented ("type D") attachment, the most severe form of attachment pathology. This attachment category is associated with early trauma, and will be presented as a prototype of maladaptive infant mental health. Throughout I shall underscore the effects of the caregiveris stress regulating and dysregulating psychobiological interactions on the infantis maturing coping systems that are organizing in the limbic circuitries of the early developing right hemisphere. An increasing body of evidence indicates that "maternal care during infancy serves to eprogrami behavioral responses to stress in the offspring" (Caldji et al., 1998. p. 5335).

And so in the first of this two part contribution I will offer an overview of an interdisciplinary perspective of development, outline connections between attachment theory, stress regulation, and infant mental health, describe the neurobiology of a secure attachment, present models of right brain, early limbic system, and orbital frontolimbic development, and suggest links between continued orbitofrontal and right brain development and adaptive mental health. In the second part of this sequenced work I will offer ideas about how early relational traumatic assaults of the developing attachment system inhibit right brain development, impair affect regulating capacities, and negatively impact infant and adult mental health. These models are presented for further experimental testing and clinical validation.

Overview of an Interdisciplinary Perspective of Development

To date, infant mental health has mostly been described in terms of the presence or absence of certain psychological functions, but it should be pointed out that these functions are, in turn, the product of biological structural systems which are organizing over the stages of infancy. Such internal systems are clearly located in the developing brain which mediates more complex functions, and it is known that the conditions and events occurring in "critical" or "sensitive" early periods of brain development have long-enduring effects. Brazelton and Cramer (1990) note that in critical phases energy is high in the infant and the parent for receptivity to each other's cues and for adapting to each other.

From late pregnancy through the second year the brain is in a critical period of accelerated growth, a process that consumes higher amounts of energy than any other stage in the life span, and so it requires sufficient amounts of not only nutrients, especially long-chain polyunsaturated fatty acids (Dobbing, 1997) but also regulated interpersonal experiences for optimal maturation (Levitsky & Strupp, 1995; Schore, 1994). The critical period concept, now firmly established in biology (Katz, 1999), prescribes that "specific critical conditions or stimuli are necessary for development and can influence development only during that period" (Erzurumlu & Killackey, 1982, p. 207). But it also suggests that during critical periods brain growth is exquisitely susceptible to adverse environmental factors such as nutritional deficits and dysregulating interpersonal affective experiences, both of which negatively impact infant mental health.

The human brain growth spurt, which is at least 5/6 postnatal, begins in the third trimester in utero and continues to about 18-to-24-months-of-age (Dobbing & Sands, 1973). During this period the brain is rapidly generating nucleic acids that program developmental processes at a rate that will never again be attained. This massive production of both nuclear and mitochondrial genetic material in the infantis brain is directly influenced by events in specifically the social-affective environment (Schore, 1994). Indeed, the most recent conceptions of development utilize a "transactional model," which views development and brain organization as "a process of transaction between (a) genetically coded programs for the formation of structures and connections among structures and (b) environmental influence" (Fox, Calkins, & Bell, 1994, p. 681). And so Sander (2000) formulates a key question for deeper understandings of infant mental health:

To what extent can the genetic potentials of an infant brain be augmented or optimized through the experiences and activities of the infant within its own particular caregiving environment? (p. 8).

The interface of nature and nurture occurs in the psychobiological interaction between mother and infant, "the first encounter between heredity and the psychological environment" (Lehtonen, 1994, p. 28). According to Cicchetti and Tucker, "Environmental experience is now recognized to be critical to the differentiation of brain tissue itself. Natureis potential can be realized only as it enabled by nurture" (1994, p. 538). The evolution and specification of this potential is described in the current biological literature:

One of the most fundamental strategies for biological adaptation in organisms is the ability of the central nervous system (CNS) to react and modify itself to environmental challenges. There is general agreement that the genetic specification of neuronal structure is not sufficient for an optimally functional nervous system. Indeed, a large variety of experimental approaches indicate that the environment affects the structure and function of the brain (Gomez-Pinilla, Choi, & Ryba, 1999, p. 1051).

A large body of evidence supports the principle that cortical and subcortical networks are generated by a genetically programmed initial overabundant production of synaptic connections, which is then followed by an environmentally-driven process of competitive interaction to select those connections that are most effectively entrained to environmental information. This parcellation, the activity-dependent fine tuning of connections and pruning of surplus circuitry, is a central mechanism of the self-organization of the developing brain (Schore, 1994; Chechik, Meilijson, & Ruppin, 1999). It is important to emphasize, however, that environmental experience can either enable or constrain the structure and function of the developing brain. In other words, early interpersonal events positively or negatively impact the structural organization of the brain and its expanding adaptive functional capacities. This clearly implies, in the broadest of terms, a direct relationship between an enabling socioemotional environment, an optimally developing brain, and adaptive infant mental health.

A major conclusion of the last decade of developmental neuroscience research is that there is now agreement that the infant brain "is designed to be molded by the environment it encounters" (Thomas et al., 1997, p. 209). The brain is thus considered to be a bioenvironmental or biosocial organ (Gibson, 1996), and investigators are now exploring the unique domains of the "social brain" (Brothers, 1990), and the central role of emotions in social communication (Adolphs, 2000). In applying this principle to social-emotional development, the connections between the neurobiological concept of "enriched environment" and the psychological concept of "optimal development" can now be more closely coupled in the psychoneurobiological construct of a "growth-facilitating" (as opposed to "growth-inhibiting") interpersonal environment (Greenspan, 1981; Schore, 1994) that positively (or negatively) effects the experience-dependent maturation of the brain.

This interdisciplinary model is compatible with very recent conceptions which emphasize that developmental processes can best be understood in terms of a context in which evolving biological systems are interacting with the social realm. As Cairns and Stoff describe:

It is necessary to go beyond the conventional notion that biological variables not only influence behavior and environment to the more modern notion that behavioral and environmental variables also impact on biology. Maturation and developmental processes may provide the common ground for understanding the process of biological social integration. On the one hand, it is virtually impossible to conceptualize developmental changes without recognition of the inevitable internal modifications that occur within the organism over time. On the other hand, it is misleading to focus on the individualis biology in the absence of detailed information about the interaction and social circumstances in which the behavior occurs (1996, p. 349).

This integration of biology and psychology in order to understand development has a rich tradition in science. In The Expression of Emotions in Man and Animals, Darwin (1872) established the scientific study of emotions and proposed that movements of expression in the face and body serve as the first means of communication between the mother and her infant (Schore, 2000a, b, c). And in The Project for a Scientific Psychology, Freud (1895), in an attempt to link neurology and psychology, first presented both his models of early development and ideas on how early traumatic events could heighten the risk of later forming psychopathology (Schore, 1995; 1997a, c). Although others have followed this line of integrating the biological and psychological realms, perhaps the most important scientist of the late twentieth century to apply an interdisciplinary perspective to the understanding of how early developmental processes influence later mental health was John Bowlby. Over two decades ago he asserted that attachment theory can frame specific hypotheses that relate early family experiences to different forms of psychiatric disorders, including the neurophysiological changes that accompany these disturbances of mental health. It is thus no coincidence that attachment theory, the dominant theoretical model of development in contemporary psychology, psychoanalysis, and psychiatry, is the most powerful current source of hypotheses about infant mental health.

Attachment, Stress Regulation, and Infant Mental Health

In his classic work of developmental science Bowlby (1969) called for deeper explorations of how an immature organism is critically shaped by its primordial relationship with a mature adult member of its species, that is, more extensive studies of how an attachment bond forms between the infant and mother (Schore, 2000a, b). In this conception, developmental processes are the product of the interaction of a unique genetic endowment with a particular "environment of adaptiveness, and especially of his interaction with the principal figure in that environment, namely his mother" (Bowlby 1969; p. 180). Thus, the infantis emerging social, psychological, and biological capacities can not be understood apart from its relationship with the mother.

More specifically, in Attachment (1969) Bowlby inquired into the mechanisms by which the infant forms a secure attachment bond of emotional communication with the mother, and how this early socioemotional learning is then internalized in the form of an enduring capacity to regulate and thereby generate and maintain states of emotional security. He observed that the mother-infant attachment relationship is "accompanied by the strongest of feelings and emotions, happy or the reverse," (p. 242), that this interaction occurs within a context of "facial expression, posture, tone of voice, physiological changes, tempo of movement, and incipient action," (p. 120), that attachment interactions allow for the emergence of a biological control system which functions in the organismis "state of arousal" (pp. 152-157), that the instinctive behavior which constitutes attachment emerges from the co-constructed environment of evolutionary adaptiveness has consequences that are "vital to the survival of the species" (p. 137), and that the infantis "capacity to cope with stress" is correlated with certain maternal behaviors (p. 344). These last two factors, adaptiveness and coping capacity are obviously central components of infant mental health.

In recent writings I have contended that attachment theory is, in essence, a regulatory theory (Schore, 2000a, b, c). More specifically, in such attachment transactions the secure mother, at an intuitive, nonconscious level, is continuously regulating the babyis shifting arousal levels and therefore emotional states. Emotions are the highest order direct expression of bioregulation in complex organisms (Damasio, 1998), and attachment can thus be defined as the dyadic regulation of emotion (Sroufe, 1996). As a result of being exposed to the primary caregiveris regulatory capacities, the infantis expanding adaptive ability to evaluate on a moment-to-moment basis stressful changes in the external environment, especially the social environment, allows him or her to begin to form coherent responses to cope with stressors. It is important to note that not just painful experiences but novel events are stressors. This means that the capacity to orient towards not only the familiar but to approach, tolerate, and incorporate novelty is fundamental to the expansion of a developing systemis adaptive capacity to learn new information and therefore to move towards more complexity.

Furthermore, since the maturation of the brain systems that mediate this coping capacity occurs in human infancy, the development of the ability to adaptively cope with stress is directly and significantly influenced by the infantis early interaction with the primary caregiver (Schore, 1994, 1997b; 2000b). In support of Bowlbyis speculations on the association of attachment with coping mechanisms, recent interdisciplinary studies indicate that "even subtle differences in maternal behavior can affect infant attachment, development, and physical well-being" (Champoux, Byrne, DeLizio, & Suomi, 1992, p. 254), and that "variations in maternal care can serve as the basis for a nongenomic behavioral transmission of individual differences in stress reactivity across generations" (Francis, Diorio, Liu, & Meaney, 1999, p. 1155).

In other words, the same interactive regulatory transactions that co-create a secure attachment bond also influence the development and expansion of the infantis regulatory systems involved in appraising and coping with stress, and therefore essential to organismic survival. According to McEwen and Stellar, "A stressful stimulus results in a severe perturbation of an organismis physiological systems, and the degree of the perceived or real threat determines the magnitude of the stress response to an internal or extrernal challenge" (1993, p. 2093). In describing stress, a concept that lies at the interface of the biological and psychological realms, Weinstock (1997) states:

The survival of living organisms depends upon the maintenance of a harmonious equilibrium or homeostasis in the face of constant challenge by intrinsic or extrinsic forces or stressors. Stress is a term that is widely used to describe both the subjective experience induced by a novel, potentially threatening or distressing situation, and the behavioral or neurochemical reactions to it. These are designed to promote adaptive response to the physical and psychological stimuli and preserve homeostasis...Successful equilibrium is reflected by a rapid neurochemical response to these stimuli which is terminated at the appropriate time, or gives way to counter-regulatory measures to prevent an excessive reaction (p. 1).

There is now agreement that the these critical functions are mediated by the sympathetic-adrenomedullary (SAM) axis and the hypothalamo-pituitary-adrenocortical (HPA) axis. Furthermore, a growing body of studies indicates that the threshold for stimulation of the SAM axis is lower than that for stimulation of the HPA axis (Malarkey, Lipkus, & Cacioppo, 1995), and that the neurochemistry of the former is regulated by the major stress hormone, corticotropin releasing factor (CRF) which regulates catecholamine release in the sympathetic nervous system (Brown et al., 1982), and of the latter by the glucocorticoid, cortisol, the major "anti-stress" hormone (Yehuda, 1999). Yehuda points out that the greater the severity of the stressor, the higher the levels of these neurochemicals, and also that the actions of these two systems are synergistic: "whereas catecholamines facilitate the availability of energy to the bodyis vital organs, cortisolis role in stress is to help contain, or shut down sympathetic activation" (1999, p. 257).

In other words, the energy-expending sympathetic and energy-conserving parasympathetic components of the autonomic nervous system (ANS) regulate the autonomic, somatic aspects of not only stress responses but emotional states. This adaptive function is stressed by Porges (1997, p. 65):

Emotion depends on the communication between the autonomic nervous system and the brain; visceral afferents convey information on physiological state to the brain and are critical to the sensory or psychological experience of emotion, and cranial nerves and the sympathetic nervous system are outputs from the brain that provide somatomotor and visceromotor control of the expression of emotion.

But in addition to the ANS, there is now a growing appreciation of the role of the central nervous system (CNS) limbic circuits in coping capacities, since this emotion-processing system is specialized to appraise social information from facial expressions implicitly, without conscious awareness (Critchley et al., 2000a), to represent motivationally salient stimuli in order to adapt to a rapidly changing environment (Mesulam, 1998), and to alter the activity of brainstem neuromodulatory systems responsible for emotional states and arousal (Tucker, 1992). These subcortically produced neuromodulatory bioamines, especially the catecholamines dopamine and noradrenaline, regulate brain state (Flicker, McCarley, & Hobson, 1981), energy metabolism (Huang et al., 1994) and blood flow microcirculation (Krimer et al., 1998). By activating cAMP-response-element-binding protein (CREB; Walton & Dragunow, 2000), they also act as internal clocks to coordinate the timing of developmental processes (Lauder & Krebs, 1986) and mediate both trophic-growth-promoting and stress related functions (Morris, Seidler, & Slotkin, 1983; Schore, 1994; OiDowd et al., 1994). The limbic system is involved in stress functions (Seyle, 1956), and various components of this system are responsible for appraising the salience of a stressor, and then initiating and organizing a psychobiological response.

Current developmental research indicates that individual differences in peripheral and central autonomic balance emerge in early development, and that these are reflected in the affective and cognitive domains (Friedman & Thayer, 1998). The "lower," subcortical sympathetic and parasympathetic components of the ANS, as well as the "higher" cortical limbic components of the CNS, are organizing pre- and postnatally, and their maturation is experience-dependent (Schore, 1996; 2000d). In fact it is now thought that

(an) early postnatal period represents a "critical period" of limbic-autonomic circuit development, during which time experience or environmental events might participate in shaping ongoing synapse formation (Rinaman, Levitt, & Card, 2000, p. 2739).

This organization is especially expressed in the early maturing (Chiron et al., 1997) right hemisphere, which, more so than the later developing left, deeply connects into both the limbic system (Tucker, 1992) and ANS (Spence, Shapiro, & Zaidel, 1996), and is therefore dominant for the human stress response (Wittling, 1997) and organismic survival (Wittling & Schweiger, 1993). The environmental events that influence ANS-limbic circuit development are embedded in the infantis ongoing affect regulating attachment transactions. Bowlby suggested that the limbic system is intimately tied to attachment, an idea furthered by Anders & Zeanah (1984). But these circuits are emphasized in specifically the right brain, because compared to the left, "the right limbic system may be better connected with subcortical neurochemical systems associated with emotion" (Buck, 1994, p. 272).

It is now accepted that in a growth-facilitating social enviornment the attachment interactions the child has with its mediators influences the maturation of connections within her developing limbic system (Schore, 1994), and that cortical paralimbic networks are formed through "ontogenetic plasticity, that is, through a natural selection of those connections that match the data in the environment" (Tucker, 1992, p. 109). On the other hand, current developmental neurobiological research reveals that growth-inhibiting, adverse early rearing experiences "have longstanding and complex effects on a range of neurochemicals relevant to emotion regulation" (Coplan et al., 1998, p. 473). Severely compromised attachment histories are thus associated with brain organizations that are inefficient in regulating affective states and coping with stress (Schore, 1997b), and therefore engender maladaptive infant mental health. This deficit is expressed in a failure to move away from homeostasis in order to turn on neurochemical stress responses when needed, and/or to turn them off and re-establish homeostasis when they are no longer needed.

As Emde (1988) has pointed out, a developmental orientation indictates that maladaptive functioning is specifically manifest as a lack of variability when an individual is faced with environmental demands that call for alternative choices and strategies for change. In light of the principle that the process of reestablishing homeostasis in the face of challenge allows for the adaptive capacity of "achieving stability through change" (Schulkin, Gold, & McEwen, 1998, p. 220), this deficit results in not just an unstable self system but one with a poor capacity to change, a limited ability to continue to develop at later points in the life cycle. Crittenden and DiLalla describe:

Adaptive development can be considered a product of the interaction of a changing biological organism with its environment such that the organism is effective in using the resources of its environment to meet its present needs without jeopardizing its future development. Maladaptive developmental courses either do not meet the organismis present needs as well as others or they reduce the organismis responsiveness to future change (1988, p. 585).

This relationship between events in early development and a later capacity for change is due to the fact that the early social environment directly impacts the experience-dependent maturation of the limbic system, the brain areas specialized for the organization of new learning and the capacity to adapt to a rapidly changing environment (Mesulam, 1998). Because limbic areas in the cortex and subcortex are in a critical period of growth in the first two years and these same neurobiological structures mediate stress-coping capacities for the rest of the the life span, early interpersonal stress-inducing and stress-regulating events have long-enduring effects.

Indeed, recent developmental psychobiological studies suggest that:

An individualis response to stressful stimuli may be maladaptive producing physiological and behavioral responses that may have detrimental consequences, or may be adaptive, enabling the individual to better cope with stress. Events experienced early in life may be particularly important in shaping the individualis pattern of responsiveness in later stages of life (Kehoe et al., 1996, p. 1435).

This conception suggests direct links between infant and adult mental health.

Integrating these conceptualizations, I suggest that adaptive infant mental health can be fundamentally defined as the earliest expression of efficient and resilient strategies for coping with novelty and stress, and maladaptive infant mental health as a deficit in these same coping mechanisms. The former is a resilience factor for coping with psychobiological stressors at later stages of the life cycle, the latter is a risk factor for interruptions of developmental processes and a vulnerability to the coping deficits that define later-forming psychopathogies. Both are attachment outcomes, and so this formulation is congruent with Mainis (1996) assertion that "disorganized" and "organized" forms of insecure attachment are primary risk factors for the development of mental disorders.

Affect Synchrony, Resonance, and Attachment Communications

The ontogeny of adaptive infant mental health is positively correlated with the ongoing development of attachment experiences over the first year. This is due to the fact that the experience-dependent maturation of the babyis brain allows for the emergence of more complex functional capacities for coping with stressors, especially those from the social environment. This developmental advance is an outcome of the co-creation of a secure attachment bond of emotional communication between infant and mother. It has been said that "learning how to communicate represents perhaps the most important developmental process to take place during infancyi (Papousek & Papousek, 1997, p. 42). What do we know about the relationships between the earliest development of socioemotional communication and the organization of adaptive brain systems?

From birth onwards, the infant is using its expanding coping capacities to interact with the social environment. In the earliest proto-attachment experiences, the infant is utilizing its maturing motor and developing sensory capacities, especially smell, taste, and touch, to interact with the social environment. As described by Trevarthen (this journal) and confirmed in very recent research on rhythmic discriminations in newborns (Ramus et al., 2000), auditory stimuli are also impacting the infantis developing sensory systems. But by the end of the second month there is a dramatic progression of its social and emotional capacities. In two functional magnetic resonance imaging (fMRI) studies, Yamada et al. (1997, 2000) demonstrate a milestone for normal development of the infant brain occurs at about 8 weeks. At this point a rapid metabolic change occurs in the primary visual cortex of infants. These authors interpret this rise to reflect the onset of a critical period during which synaptic connections in the occipital cortex are modified by visual experience.

With this maturational advance, the visual stimuli emanating from the mother's emotionally expressive becomes the most potent stimulus in the infant's social environment, and the child's intense interest in her face, especially in her eyes, leads him to track it in space, and to engage in periods of intense mutual gaze. The infant's gaze, in turn, evokes the mother's gaze, thereby acting as a potent interpersonal channel for the transmission of "reciprocal mutual influences." In the developmental psychological literature Fogel and Branco (1997) characterize infant emotional metacommunication in parent-infant interaction expressed in nonverbal gaze direction, facial expression, posture, and body movements that are "mutually coordinated to create emergent social patterns" (p. 68). And writing in the neurobiological literature, Allman and Brothers assert, "When mutual eye contact is established, both participants know that the loop between them has been closed...and this is the most potent of all social situations" (1994, p. 61).

In very recent basic research on three-month-old infants, Feldman, Greenbaum, and Yirmiya (1999, p. 223) describe:

Face-to-face interactions, emerging at approximately 2 months of age, are highly arousing, affect-laden, short interpersonal events that expose infants to high levels of cognitive and social information. To regulate the high positive arousal, mothers and infants...synchronize the intensity of their affective behavior within lags of split seconds.

These episodes of "affect synchrony" occur in the first expression of social play, and at this time they are patterned by an infant-leads-mother-follows sequence. This highly organized dialogue of visual and auditory signals is transacted within milliseconds, and is composed of cyclic oscillations between states of attention and inattention in each partneris play. In this interactive matrix both partners match states and then simultaneously adjust their social attention, stimulation, and accelerating arousal to each otheris responses.

Feldman and her colleagues assert,

Synchronicity is defined as a match between motheris and infantis activities that promotes positivity and mutuality in play. By synchronizing with the childis attentive states, mothers structure playful interactions, regulate infant attention, facilitate the development of verbal dialogue, and promote the infantis capacity for self-regulation...mutual synchrony exists when both partners simultaneously adjust their attention and stimulation in response to the partneris signals (1996, p. 349).

These are critical events, because they represent a fundamental opportunity to practice the interpersonal coordination of biological rhythms. According to Lester, Hoffman, and Brazelton "synchrony develops as a consequence of each partneris learning the rhythmic structure of the other and modifying his or her behavior to fit that structure" (1985, p. 24).

In this process of "contingent responsivity," not only the tempo of their engagement but also their disengagement and reengagement is coordinated. The more the psychobiologically attuned mother tunes her activity level to the infant during periods of social engagement, the more she allows him to recover quietly in periods of disengagement, and the more she attends to the childis reinitiating cues for reengagement, the more synchronized their interaction. The period immediately after a "moment of meeting," when both partners disengage, provides "open space," in which both can be together, yet alone (autoregulating) in the presence of the other (Sander, 1988). The synchronizing caregiver thus facilitates the infant's information processing by adjusting the mode, amount, variability, and timing of the onset and offset of stimulation to the infant's actual integrative capacities. These mutually attuned synchronized interactions are fundamental to the healthy affective development of the infant (Penman, Meares, & Milgrom-Friedman, 1983).

In these exchanges of affect synchrony, as the mother and infant match each other's temporal and affective patterns, each recreates an inner psychophysiological state similar to the partner's. Stern (1983b) describes moment-to-moment state sharing, feeling the same as the other, and state complementing, responding in oneis unique way to stimuli coming from the other. In contexts of "mutually attuned selective cueing", the infant learns to preferentially send social cues to which the mother has responded, thereby reflecting "an anticipatory sense of response of the other to the self, concomitant with an accommodation of the self to the other" (Bergman, 1999, p. 96).

In describing the unique nature of an emotionally communicationg mother-infant dyad, a number of prominent theoreticians have been drawn to the concept of resonance. Trevarthen describes

Corresponding generative parameters in...two subjects enable them to resonate with or reflect on one another as minds in expressive bodies. This action pattern can become eentrained,i and their experiences can be brought into register and imitated. These are the features that make possible the kind of affectionate empathic communication that occurs, for instance, between young infants and their mothers (1993, p. 126).

Simliarly, Sander (1991) emphasizes the critical importance of the context of a specifically fitted interaction between the infant and mother as a resonance between two systems attuned to each other by corresponding properties. Such energy-infused moments allow for a sense of vitalization, and thereby increased complexity and coherence of organization within the infant.

Furthermore, in the visual and auditory emotional communications embedded within synchronized face-to-face transactions both members of the dyad experience a state transition as they move together from low arousal to a heightened energetic state of high arousal, a shift from quiet alertness into an intensely positive affective state. In physics, a property of resonance is sympathetic vibration, which is the tendency of one resonance system to enlarge and augment through matching the resonance frequency pattern of another resonance system. It is well established that energy shifts are the most basic and fundamental features of emotion, that the transfer of emotional information is intensified in resonant contexts, and that at the moment when a system is tuned at the "resonant" frequency it becomes synchronized (Schore, 1997b, 2000d; in press a).

Resonances often have chaos associated with them, and thus they are characterized by non-linear dynamical factors - relatively small input amplitudes engender a response with a surprisingly large output amplitude. This amplification especially occurs when external sensory stimulation frequency coincides with the organismis own endogenous rhythms. In other words, when a psychobiologically attuned dyad co-creates a resonant context within an attachment transaction, the behavioral manifestation of each partneris internal state is monitored by the other, and this results in the coupling between the output of one partneris loop and the input of the otheris to form a larger feedback configuration and an amplification of the positive state in both.

In demonstration of this principle, emotion theorists describe "affect bursts," nonverbal expressions of synchronized facial and vocal activity triggered by an external stimulus (Scherer, 1994). And infant researchers refer to the delight the infant displays in reaction to the augmenting effects of his motheris playful, empathically attuned behavior, her mulitmodal sensory amplification and resonance with the childis feelings. Stern (1985) describes a particular maternal social behavior which can "blast the infant into the next orbit of positive excitation," and generate "vitality affects." In these transactions the dyad is co-creating "mutual regulatory systems of arousal" (Stern, 1983a).

In this interactive context, the infantis attachment motivation synergistically interacts with the caregiveris maternal motivation. In current psychobiological models maternal motivation is conceptualized as the outcome of the interaction between external visual and auditory infant stimuli and the central state of maternal arousability (Pryce, 1992). In order to act as a regulator of the infantis arousal, she must be able to regulate her own arousal state. The burgeoning capacity of the infant to experience increasing levels of accelerating, rewarding arousal states is thus at this stage amplified and externally regulated by the psychobiologically attuned mother, and depends upon her capacity to engage in an interactive emotion communicating mechanism that generates these in herself and her child.

Reciprocal facial signalling, mutual rhythmic entrainment, and dyadic resonance thus act as a psychobiological context for an open channel of social communication, and this interactive matrix promotes the outward expression of internal affective states in infants. Sander (1997) asserts that the parent expresses a behavior that is particularly fitted to catalyze a shift in the infantis state, and Tronick et al. (1998) state that the complexity of the infantis state is expandable with input from an external source - the caregiver. In order to enter into this communication, the mother must be psychobiologically attuned not so much to the child's overt behavior as to the reflections of the rhythms of his internal state.

Since affect attunements are "spontaneous, nonverbal responses to...childrenis expressed emotions" (Polan & Hofer, 1999, p. 176), the moment-to-moment expressions of the motheris regulatory functions occur at levels beneath awareness. Even so, the attuned mother can self-correct by accessing her reflective function whereby she monitors not only her infantis but her own internal signals and differentiates her own affective state. As a regulator of the infantis arousal levels, she also modulates nonoptimal high levels of stimulation which would induce supra-heightened levels of arousal in the infant. Thus she regulates not just the type but also the intensity of socioaffective information within the dyadis communication system.

But the primary caregiver is not always attuned - developmental research shows frequent moments of misattunement in the dyad, ruptures of the attachment bond. In early development an adult provides much of the necessary modulation of infant states, especially after a state disruption and across a transition between states, and this allows for the development of self regulation. Again, the key to this is the caregiver's capacity to monitor and regulate her own affect, especially negative affect. The regulation of her own affective state, as well as the childis, may be an emotionally demanding task.

In this essential regulatory pattern of "disruption and repair" (Beebe & Lachmann, 1994; Schore, 1994) the "good-enough" caregiver who induces a stress response in her infant through a misattunement, reinvokes in a timely fashion her psychobiologically attuned regulation of the infant's negative affect state that she has triggered. The reattuning, comforting mother and infant thus dyadically negotiate a stressful state transition of affect, cognition, and behavior. This recovery mechanism underlies the phenomenon of "interactive repair" (Tronick, 1989; Lewis, 2000), in which participation of the caregiver is responsible for the reparation of stressful dyadic misattunements.

If attachment is interactive synchrony, stress is defined as an asynchrony in an interactional sequence, but a period of synchrony following this allows for stress recovery (Chapple, 1970). It is now thought that the process of reexperiencing positive affect following negative experience may teach a child that negativity can be endured and conquered. Infant resilience emerges from an interactive context in which the child and parent transition from positive to negative and back to positive affect, and resilience in the face of stress is an ultimate indicator of attachment capacity and therefore adaptive mental health.

These arousal-regulating transactions, which continue throughout the first year, underlie the formation of an attachment bond between the infant and primary caregiver. An essential attachment function is "to promote the synchrony or regulation of biological and behavioral systems on an organismic level" (Reite & Capitanio, 1985, p. 235). Indeed, psychobiological attunement and the interactive mutual entrainment of physiological rhythms are fundamental processes that mediates attachment bond formation, and attachment can be defined as the regulation of biological synchronicity between organisms (Wang, 1997; Schore, 2000b). The mechanism of attachment dynamics is thus an example of the regulation of rhythm, which is a fundamental organizing principle of all living systems (Iberall & McCulloch, 1969).

To put this another way, the infantis developing regulatory and control systems create spontaneous physiological rhythms that are manifest in arousal fluctuations, which are in turn expressed in fluctuating psychobiological affective states, what Stern (1985) calls vitality affects. It is now accepted that affects reflect an individualis internal state and have an hedonic (valenced) dimension and an arousal (intensity) dimension. The crescendos and decrescendos of the infantis peripheral (ANS) and central (CNS) arousal systems underlie emotions, and so the mutual entrainment of affective states in attachment transactions can be defined as the dyadic regulation of emotion (Sroufe. 1996). Thus Damasio (1998) is correct in characterizing emotions as the highest order direct expression of bioregulation in complex organisms, but it should be emphasized that the efficient bioregulation of internal emotional states can take the form of both interactive regulation and autoregulation.

These data underscore an essential principle overlooked by many emotion theorists - affect regulation is not just the reduction of affective intensity, the dampening of negative emotion. It also involves an amplification, an intensification of positive emotion, a condition necessary for more complex self-organization. Attachment is not just the restablishment of security after a dysregulating experience and a stressful negative state, it is also the interactive amplification of positive affects, as in play states. Regulated affective interactions with a familiar, predictable primary caregiver create not only a sense of safety, but also a positively charged curiosity that fuels the burgeoning selfis exploration of novel socioemotional and physical environments (Schore, 1994; Grossman, Grossman, & Zimmerman, 1999). This ability is a marker of adaptive infant mental health.

Attachment and the Interactive Regulation of the Right Brain

In a number of contributions I have offered evidence which indicates that the emotional communications of evolving attachment transactions directly impact the experience-dependent maturation of the infantis developing brain. Trevarthen (1993) also observes that that the growth of the baby's brain literally requires brain-brain interaction and occurs in the context of a positive affective relationship (see Figure 1). But in light of the fact that the early maturing right hemsphere is in a growth spurt in the first year-and-a-half, and that it is dominant for the first three (Chiron et al., 1997), I have contended that attachment experiences specifically impact the development of the infantis right brain. Confirming this model, Ryan, Kuhl, and Deci, using EEG and neuroimaging data, now report,

The positive emotional exchange resulting from autonomy-supportive parenting involves participation of right hemispheric cortical and subcortical systems that participate in global, tonic emotional modulation (1997, p. 719).

In an elegant phrase Trevarthen asserts that "the intrinsic regulators of human brain growth in a child are specifically adapted to be coupled, by emotional communication, to the regulators of adult brains" (Trevarthen, 1990, p. 357). But again, I would amend this general statement to suggest that the regulators of both the infant and motheris brains are located in specifically the right limbic brain (Schore, 1994). Furthermore, Trevarthenis description of "emotional communication" as a traffic of visual, prosodic auditory, and gestural signals that induce instant emotional effects is paralleled by Buckis (1994) characterization of "spontaneous emotional communication":

Spontaneous communication employs species-specific expressive displays in the sender that, given attention, activate emotional preattunements and are directly perceived by the receiver...The emeaningi of the display is known directly by the receiver...This spontaneous emotional communication constitutes a conversation between limbic systems...It is a biologically-based communication system that involves individual organisms directly with one another: the individuals in spontaneous communication constitute literally a biological unit...The direct involvement with the other intrinsic to spontaneous communication represents an attachment that may satisfy deeply emotional social motives (p. 266, my italics).

Buck (1994) emphasizes the importance of the right limbic system, and localizes this biologically-based spontaneous emotional communication system to the right hemisphere, in accord with other research that indicates a right lateralization of spontaneous gestures (Blonder et al., 1995) and emotional communication (Blonder, Bowers, & Heilman, 1991).

Recall Winnicottis (1971) description of the infantis expression of a "spontaneous gesture," a somato-psychic expression of the burgeoning "true self," and the attuned motheris "giving back to the baby the babyis own self." Winnicott contends that as a result of its transactions with the mother, the infant, through identification, internally creates a "subjective object." Recent research indicates that the right hemisphere is specialized for "the detection of subjective objects" (Atchley & Atchley, 1998, and for the processing and regulation of self-related information (Schore, 1994; Ryan et al., 1997; Keenan, Wheeler, Gallup, & Pascual-Leone, 2000).

Furthermore, developmental neuroscientists have proposed that engrams related to emotional voices are more strongly imprinted into the early maturing, more active right hemisphere (Carmon & Nachson, 1973), and that particular areas of the right hemisphere are timed to be in a plastic and receptive state at the very time when polysensory information that emanates from faces is being attended to most intensely by the infant (Deruelle & de Schonen, 1998; de Schonen, Deruelle, Mancini, & Pascalis, 1993). These latter authors report that right hemisphere activation in face processing shows a significant structural advance at 2-3 months, in line with the previously cited work of Yamada et al. (1997, 2000) and Feldman et al. (1999). With ongoing episodes of affective synchrony, attachment functions mature later in the first year, and it has been suggested that "there is earlier maturation of right hemisphere inhibition over subcortically mediated emotional expressions in infancy, once cortical influences over this behavior come into play" (Best & Queen 1989, p. 273).

An accumulating body of evidence indicates that the infantis right hemisphere is involved in attachment and the motheris right hemisphere in comforting functions (Henry, 1993; Horton, 1995; Schore, 1994; 1998a, b; 1999d; Shapiro, Jamner, & Spence, 1997; Siegel, 1999; Wang, 1997). Attachment represents the regulation of biological synchronicity between organisms, and imprinting, the learning process that mediates attachment, is defined as synchrony between sequential infant-maternal stimuli and behavior (Petrovich & Gewirtz, 1985). During the sequential signalling of play epsiodes mother and infant show sympathetic cardiac acceleration and then parasympathetic deceleration in response to the smile of the other (Donovan, Leavitt, & Balling, 1978). Imprinting is thus not a unidirectional learning process by which attachment experiences are passively absorbed into an empty template. Rather it is an active dyadic process that occurs between two brains that are co-generating synchronized emotional communications with each other.

I suggest that when two right brain systems are mutually entrained in affective synchrony they create a context of resonance, which is now thought to play a fundamental role in brain organization, CNS regulatory processes, and the organization of connectivity properties that are tuned by function (Salansky, 1998). Earlier I described how in face-to-face contexts resonant amplification occurs when the frequency patterns of the motheris exogenous sensory stimulation coincides with the infantis own endogenous organismic rhythms. Trevarthen (1993) points out that the resonance of the dyad ultimately permits the intercoordination of positive affective brain states.

In current neuroscience resonance refers to the ability of neurons to respond selectively to inputs at preferred frequencies, and "amplified resonance" or "amplifying currents" serve as a substrate for coordinating (synchronizing) patterns of network (circuit) activity. Basic research establishes that different behavioral and perceptual states are associated with different brain rhythms, that a resonant system evolves continuously into a spontaneously oscillatory system as the amplifying conductance is increased, and that amplified resonance can "tune networks to operate in frequency ranges of special biological meaning" (Hutcheon & Yarom, 2000, p. 220).

These general principles apply to face-to-face transactions, where patterns of information emanating from the caregiveris face, especially of low visual and auditory frequencies are specifically processed by the infantis right hemisphere (Ornstein, 1997). The ventral stream (Ungerleider & Haxby, 1994) of this hemisphere is specialized to analyze low frequencies of visual perception that convey the general outlines of faces and low frequencies of auditory tones that express the emotional intonation of language, prime examples of biologically meaningful information. Fernald (1992) describes human maternal vocalizations to infants as "biologically relevant signals." Furthermore, these dyadically synchronized affectively charged transactions elicit high levels of metabolic energy for the tuning of right brain cortical-subcortical circuits involved in processing socioemotional information (Schore, 1994, 1997b, 2000d). An article in Science suggests "mothers invest extra energy in their young to promote larger brains" (Gibbons, 1998, p. 1346).

Lewis (1995) points out that the best example of the flowthrough of energy in a developing system is the processing of relevant information in the presence of emotion. Thus, as a result of synchronized emotional transactions, the organization of the infantis right brain shows increased coherence, as the flow of energy between the hierarchically organized higher right cortical and lower right subcortical components increase their connectivity, allowing the right brain to act as a self-regulating integrated whole, and therefore capable of increasing complexity. This conception is consonant with current models that emphasize that the brain is a self-organizing system (van Pelt et al., 1994), and that age increases brain complexity (Anokhin et al., 1996). In applying dynamic systems principles to attachment theory, Siegel (1999) proposes a similiar scenario.

The infantis right brain is tuned to dynamically self-organize upon perceiving certain patterns of facially expressed exteroceptive information, namely the visual and auditory stimuli emanating from the smiling and laughing joyful face of a loving mother. In face-to-face interactive affect-amplifying transactions, the relational context triggers facially expressed "affect bursts" in the infant. According to Scherer, these highly emotionally charged events lead to a "strong synchronization of various organismic subsystems, particularly the various expressive channels, over a very brief period of time" (1994, p. 181).

What psychoneurobiological mechanism could underlie this caregiver-induced organization of the infantis brain? In earlier work I have suggested that the appearance of the motheris face in dyadic play experiences generates high levels of dopaminergic-driven arousal and elation in the infantis right brain (Schore, 1994). Dopamine neurons in the ventral tegmental area of the anterior reticular formation are involved in reward and emotionality (Wise & Rompre, 1989), and they respond to visual, auditory, and tactile stimuli by switching from "pacemaker-like firing" to "burst firing" (Gonon, 1988; Overton & Clark, 1997) in response to an environmental stimulus that is "ethologically salient" (a good definition of sensory stimulation emanating from the mother). This pacemaker firing of a subnuclei of arousal-generating ventral tegmental dopamine neurons may represent an important component of the infantis genetically encoded endogenous organismic rhythms.

The bursting of these neurons to salient, arousing environmental stimuli contributes to an orienting response, the setting of a motivational state, and the onset of exploratory behavior (Horvitz, Stewart, & Jacobs, 1997). Furthermore, "electrical coupling among bursting dopamine neurons may provide a mechanism for further amplification of the effects of synchronously firing dopamine cells on their target areas" (Freeman, Meltzer, & Bunney, 1985, p. 1993). Evidence also indicates that the evaluation of an environmental stimulus as affectively positive is associated with dopaminergic activation of specifically the right brain (Besson & Louilot, 1995).

An integration of these data may give us a model of the critical right brain events by which psychobiologically attuned attachment communications generate amplified resonance that tunes reward circuits to certain forms of human visual and auditory patterns of stimulation. In affectively charged face-to-face transactions, the biologically significant information that emanates from the motheris face is imprinted into the infantis developing right inferior temporal areas that process familiar faces (Nakamura et al., 2000), and thereby takes on "special biological meaning." The right hemisphere is also dominant for the perception of "biological motion" (Grossman et al., 2000). These psychoneurobiological events of mother-infant play sequences drive the "affective bursts" embedded within moments of affective synchrony, in which positive states of interest and joy are dyadically amplified. Panksepp (1998) contends that "play may have direct trophic effects on neuronal and synaptic growth in many brain systems" (p. 296), and suggests that play serves the adaptive role of organizing affective information in emotional circuits, a function also performed by rapid eye movement (REM) dream sleep. This fits nicely with current neuroscience conceptions of the important role of REM sleep in brain maturation (Marks et al., 1995) and imaging studies showing a preferential activation of limbic regions in REM sleep (Braun et al., 1997; Maquet et al., 1996).

How can we account for the trophic effects of early play episodes? Again, in a previous contribution (Schore, 1994) I have proposed that in these face-to-face emotional communications, the visual input of the motheris face is also inducing the production of neuorotrophins in the infantis brain, such as brain-derived neurotrophic factor (BDNF). Maternal care has been shown to increase N-methyl-D-aspartate (NMDA) receptor levels, resulting in elevated BDNF and synaptogenesis in the infantis brain (Liu et al., 2000). This trophic factor, which is regulated by visual input (Gomez-Pinilla et al., 1999), promotes synaptic plasticity during postnatal critical periods (Huang et al., 1999). BDNF is also a growth-promoting factor for mesencephalic dopamine neurons (Hyman et al., 1991), and dopamine, which activates NMDA receptors (Knapp, Schmidt, & Dowling, 1990), is known to perform a growth-promoting role in the postnatal development of the cortex (Kalsbeek et al., 1987), especially in corticolimbic areas that send axons down to the dendrites of these dopamine neurons, and thereby come to regulate their activity (Sesack & Pickel, 1992; Schore, 1994). Dopamine acts as a trophic agent via regulation of the developing blood brain barrier (Schore, 1994) and microcirculation (Krimer et al., 1998) of developing target areas.

Other psychobiological data may explicate the mechanisms that mediate attachment, the interactive regulation of biological synchronicity between organisms. Despite the intrinsic dyadic nature of the attachment concept, hardly any research has concurrently measured mother and infant in the process of interacting with each other. In one of the few studies of this kind, Kalin, Shelton and Lynn (1995) show that the intimate contact between the mother and her infant is mutually regulated by the reciprocal activation of their opiate systems - elevated levels of beta endorphins increase pleasure in both brains. It is established that opioids enhance play behavior (Schore, 1994) and that endorphins increase the firing of mesolimbic dopamine neurons (Yoshida et al., 1993).

Furthering these ideas, the developmental principle of "reciprocal mutual influences" refers to more than mutual behavior changes, indeed it specifically implies that there are simultaneous changes within the right brains of both members of the dyad. In terms of self-organization theory, the mutual entrainment of their right brains during moments of affect synchrony triggers an amplified energy flow which allows for a coherence of organization that sustains more complex states within both the infantis and the motheris right brains. In this manner, "the self-organization of the developing brain occurs in the context of a relationship with another self, another brain" (Schore, 1996, p. 60).

Evidence is now appearing that supports the idea that the organization of the motheris brain is also being influenced by these relational transactions. A neurobiological study of early mammalian mother-infant interactions, published in Nature, entitled "Motherhood improves learning and memory," reports increased dendritic growth in the motheris brain (Kinsley et al., 1999). The authors conclude that events in late pregnancy and the early postpartum period

...may literally reshape the brain, fashioning a more complex organ that can accomodate an increasingly demanding environment...To consider the relationship of a mother caring for her young as unidirectional disregards the potentially rich set of sensory cues in the opposite direction that can enrich the motheris environment. By providing such stimuli, (infants) may ensure both their own and their motheris development and survival (p. 137).

Hoferis (1990) developmental psychobiological work also emphasizes the bidirectional brain events of the mother-infant interaction. He describes, in detailed fashion, how the infant's immature and developing internal homeostatic systems are co-regulated by the caregiver's more mature and differentiated nervous system. In this "symbiotic" pleasurable state, the adult's and infant's individual homeostatic systems are linked together in a superordinate organization which allows for "mutual regulation of vital endocrine, autonomic, and central nervous systems of both mother and infant by elements of their interaction with each other" (Hofer, 1990, p. 71)

These matters bear upon the concept of symbiosis, which has had a controversial history in recent developmental psychoanalytic writings. This debate centers around Mahler, Pine, and Bergmanis (1975) reference to a normal symbiotic phase during which the infant "behaves and functions as though he and his mother were a single omnipotent system-a dual unity within one common boundary" (p. 8). Although the symbiotic infant is dimly aware that the mother is the source of his pleasurable experiences, he is in a "state of undifferentiation, a state of fusion with the mother, in which the eIi is not differentiated from the enot-Ii" (p. 9).

This latter defintion of symbiosis departs from the classical biological concept and is unique to psychoanalytic metapsychology. Current evidence may not directly support any inferences about the limits of the infantis awareness, nor about an entire stage that describes the infantis behavior only with this characterization. However, moments of face-to-face affective synchrony do begin at 2-3 months, the advent of Mahleris symbiotic phase, they do generate high levels of positive arousal, and such mutually attuned sequences can be portrayed as what Mahler et al. (1975) call instances of "optimal mutual cueing."

But even more importantly, Hoferis work as well as recent brain research calls for a return of the definition of symbiosis to its biological origins. The Oxford dictionary offers the derivation from the Greek, "living together," and defines symbiosis as an interaction between two dissimilar organisms living in close physical association, especially one in which each benefits the other (my italics). An even more basic definition from biological chemistry suggests that "symbiosis is an association between different organisms that leads to a reciprocal enhancement of their ability to survive" (Lee et al., 1997, p. 591). Recall Buckis (1994) description of an emotionally communicating dyad as "literally a biological unit," a conception that echoes Polan and Hoferis (1999) description of the dyad as a self-organizing regulatory system composed of mother and infant as a unit. These conceptions suggest that instances of secure attachment bonding are an example of biological symbiosis. Interestingly, the Oxford dictionary also defines symbiosis as "companion," which suggests that Trevarthenis concept refers to this same psychobiological phenomenon.

The construct of symbiosis is reflected in the conception of attachment as the interactive regulation of biological synchronicity between organisms. In discussing the central role of facial signalling in attachment, Cole asserts, "It is through the sharing of facial expressions that mother and child become as one. It is crucial, in a more Darwinian biological context, for the infant to bond her mother to ensure her own survival " (1998, p. 11, my italics). Recall Bowlbyis (1969) assertion that the development of attachment has consequences that are vital to survival and that the infantis capacity to cope with stress is correlated with certain maternal behaviors. The right hemisphere is dominant for both attachment functions and for the control of vital functions supporting survival and enabling the organism to cope actively and passively with stress. These capacities are surely critical indices of adaptive infant mental health.

Attachment Transgresions & the Hierarchical Organization of the Limbic System

Mary Main concludes that "The formation of an attachment to a specified individual signals a quantitative change in infant behavioral (and no doubt also brain ) organization" (1991, p. 214, my italics). As a result of advances in the "decade of the brain" can we now identify what specific brain areas mediate this function? In his initial outline of attachment theory, Bowlby speculated that a "succession of increasingly sophisticated systems" involving the limbic system and brain arousal-regulating areas mediate attachment processes (1969, p. 154). It is well established that regions of the brain mature in stages, so the question is, what parts of the postnatally developing brain are maximally impacted by emotionally-charged attachment experiences? As previously mentioned the emotion processing limbic system has been implicated in attachment functions.

Indeed the first 18 months of human life are critical for the myelination and therefore the maturation of particular rapidly developing limbic and cortical association areas and limbic areas of the human cerebral cortex show anatomical maturation at about 15 months. It has long been thought that the limbic system is fundamentally associated with emotional functions. But as I stated previously, recent conceptions emphasize that limbic system function underlies the organization of new learning and the capacity to adapt to a rapidly changing environment (Mesulam, 1998). This concept relates to Hindeis assertion that "the development of social behavior can be understood only in terms of a continuing dialectic between an active and changing organism and an active and changing environment" (1990, p. 162)

Within the first year perhaps no organismic system is changing as rapidly as the brain, especially a sequence of ontogenetically appearing limbic circuits. These systems are organized from the simplest to the most complex, and they onset in a fixed progression over the first year, with the later maturing hierarchical cortical structures adaptively regulating the earlier maturing subcortical systems. This general ontogentic principle is articulated by Werner (1948), who suggested that "the development of biological forms is expressed in an increasing differentiation of parts and an increasing subordination, or hierarchization...an ordering and grouping of parts in terms of the whole (1948, p. 44). This hierarchical model has been significantly advanced in the psychoanalytic literature in the groundbreaking work of Gedo (1999; Gedo & Wilson, 1993).

In the current neuroscience literature Toates describes the importance of hierarchical control systems in development:

Development is associated with gaining autonomy from sensory control and acquisition of top-down control over behavior that is organized at a lower level. Reflexes can become integrated into cortical control. Such control will be perhaps most usually inhibition, but excitation might also occur...acquisition of higher-level control is not merely a process of more inhibition being exerted since the new forms of reacting to the environment also emerge and it is assumed that these are mediated at the higher level (1998, p. 73).

In classical ego psychology psychoanalytic writings, Hartmann (1939) proposed that adaptation is primarily a reciprocal relationship of the organism and its environment, and that development is a differentiation in which primitive regulatory systems are increasingly replaced or supplemented by more effective regulatory systems. The progression and reorganization of the infantis regulatory, control systems is described by Brazelton and Cramer:

The central nervous system, as it develops, drives infants towards mastery of themselves and their world. As they achieve each level of mastery, they seek a kind of homeostasis, until the nervous system presses them on to their next level. Internal equilibrium is always being upset by a new imbalance created as the nervous system matures. Maturation of the nervous system, accompanied by increasing differentiation of skills, drives infants to reorganize their control systems (1990, p. 98).

Fischer and Rose (1994) conclude that the development of higher order control system allows for the emergence of "dynamic skills," that a developmental stage is a point at which a new level of control systems emerge, and that emotions fundamentally shape the ways that control systems develop.

These control systems can now be identified. In current neuroscience, the neuroanatomy of the limbic system is characterized as a hierarchical system of vertically organized circuits within the brain (see Figure 2). And so authors are now referring to the "rostral limbic system" a hierarchical sequence of interconnected limbic areas in orbitofrontal, insular cortex, anterior cingulate, and amygdala (Devinsky, Morrell, & Vogt, 1995), an "anterior limbic system" composed of orbitofrontal cortex, basal forebrain, amygdala, and hypothalamus (Schnider & Ptak, 1999), a "paralimbic circuit" containing orbitofrontal, insular, and temporopolar cortices (Mesulam & Mufson, 1982), an "anterior limbic prefrontal network" interconnecting the orbital and medial prefrontal cortex with the temporal pole, cingulate, and amygdala (Carmichael & Price, 1995), and a complex circuit of emotion regulation consisting of orbital frontal cortex, anterior cingulate, and amygdala (Davidson, Putnam, & Larson, 2000).

A body of evidence shows that the orbitofrontal-insula, medial frontal anterior cingulate, and amygdala systems all interconnect with each other and with brainstem bioaminergic neuromodulatory and hypothalamic neuroendocrine nuclei (see Figure 3). Although each has reciprocal connections with dopamine neurons in the ventral tegmental area of the anterior reticular formation and noradrenaline neurons of the caudal reticular formation, each limbic subsystem maintains connections with different monoaminergic subnuclei (Halliday & Tork, 1986; Halliday et al., 1988).

Because they are all components of the limbic system, each processes and imprints a positive or negative hedonic charge on current exteroceptive information about changes in the external social environment and then integrates it with interoceptive information about concurrent alterations in internal bodily states. Due to the facts that they each directly interconnect with the ANS (Neafsey, 1990) and that autonomic activity is controlled by multiple integrative sites within the CNS that are heirarchically organized (Lane & Jennings, 1995), all are involved in the regulation of bodily-driven affective states. Although all components process exteroceptive and interoceptive information, the later maturing systems in the cortex will process this information in a more complex fashion than the earlier subcortical components. The output of the lowest level limbic levels have the character of automatic innate reflexes, while higher processing produces more flexible intuitive responses that allow fine adjustment to environmental circumstances.

In optimal socioemotional environments, each limbic level has bidirectional connections with the others, and in this manner information can both be forwarded up and down the limbic axis for further appraisal and hierarchical modulation. The earliest and simplest appraisals of exteroceptive and interoceptive affective stimuli would be hedonic and aversive affective core processes in the amygdala (Berridge, 2000), the later and most complex subjective experiences of pleasure and pain in the orbitofrontal areas (Blood et al., 1999; Francis et al., 1999; Petrovic et al., 2000). These operations are primarily lateralized to the right limbic system, which is preferentially connected downward to the right neurochemical systems associated with emotion (Buck, 1994) and upward to the ipsilateral right neocortex (Wilson et al., 1991).

The concept of a hierarchically organized brain that develops through an increasingly complex coordination of lower and higher levels was first introduced by the British neurologist Hughlings Jackson at the end of the nineteenth century. Jackson conceived of three levels of organization, including the lowest and most primitive, middle, and last to evolve, highest centers. Each of these levels is a representing system, with the highest level of integration and coordination dependent upon prefrontal activity that allows the organism as a whole to adjust to the environment (Jackson, 1931). A similar trilevel model is also seen in MacLeanis (1990) triune brian. As applied to the developmental organization of the right limbic system of the right brain, this conception suggests a three-tiered self-organizing dynamic system. Increased interconnectivity (energy flow) among the three component circuits would allow for information stored at one level to be transferred to the others. The top level that receives feedback from the lower performs an executive function (Toates, 1998), and this allows for emergent properties, that is, novel combinations of more complex emotional states.

In line with the Jacksonian ontogenetic concept of vertical brain organization (Luu & Tucker, 1996) and the principle of caudal to rostral brain development, a model of the ontogeny of the limbic system can be offered. Keeping in mind that in humans this development continues postnatally, reversing the sequence of the rostral limbic system (amygdala, anterior cingulate, insular-orbitofrontal) could offer specific ideas about how a number of discrete limbic components could come on line in a defined sequence in the first year. Recall Bowlbyis speculation that the limbic system is centrally involved in attachment and that the "upgrading of control during individual development from simple to more sophisticated is no doubt in large part a result of the growth of the central nervous system" (1969, p. 156).

The following sequence represents Bowlbyis "succession of increasingly sophisticated systems" that mediates attachment development. I further propose that the ontogenetic progression of each of these limbic subsystems progresses from an initial sympathetic-dominant excitatory phase followed by a latter parasympathetic-dominant inhibitory phase and ultimately excitation-inhibition balance (see Schore, 1994).

At birth only the amygdala (see Figure 2), a primitive limbic regulatory system that appraises crude information about external stimuli and modulates autonomic and arousal systems, is on line (Chugani, 1996). The right amygdala is known to be implicated in the processing of olfactory stimuli (Zald, Lee, Fluegel, & Pardo, 1998) within the mother and the perinatal infant relationship (Van Toller & Kendal-Reed, 1995). This suggests that right amygdala-driven processes underlie the infantis recognition of the motheris scent as well as the motheris recognition of neonates through olfactory cues (Porter, Cernoch, & McLaughlin, 1983).

Amygdala memorial systems also mediate the organization of the earliest representations of the infant maternal relationship that allow 6-day-old infants to discriminate the scent of their mother's breast pad (MacFarlane, 1977) or axillary odor (Cernoch & Porter, 1985) from that of another woman. This early appearing subcortical limbic control system is thus a central component of the proto-attachment mechanisms that are driven by the unique salience of olfactory signals (Porter & Winberg, 1999). These data further suggest that dyadic "human olfactory communication" (Russell, 1976) occurs between the motheris and infantis right brains. Limbic areas of the right hemisphere are also centrally involved in human gustation (Small et al., 1999).

The fact that the processing of olfactory/gustatory information is dominant in the perinatal period is also documented by developmental researchers. The primary organ of the body that specializes in the latter function is, of course, the mouth. According to Hernandez-Reif et al. (2000, p. 205):

The infant mouth, including the tongue, is a highly specialized multifunctional sensory-motor system designed to receive nutrients and to express discomfort, such as by varied cries that relay hunger or pain (van den Boom & Gravehurst, 1995)...Recent research revels that the newbornis mouth is also a well-developed perceptual organ. Upon contacting a non-nutritive object, the tactile receptors of the mouth generate positive presssure, presumably for the purpose of detecting object information (Butterworth & Hopkins, 1988; Rochat, 1983).

These authors point out that newborns exhibit a transfer of learning from information detected by the mouth to other sensory modalities, citing studies which show that newborns suck harder in order to see the visual (Walton, Bower, & Bower, 1992) and hear the auditory stimulation (DeCasper & Fifer, 1980) emanating from the motheris face. They also mention that at a later point of development (the second quarter of the first year - Morange-Majoux, Cougnot, & Bloch, 1997), exploration of objects shifts from the mouth to the hands (Bloch, 1998).

In the cerebral hemispheres only the primary somatosensory cortex is metabolically active at birth (Chugani, 1996), and this area is known to process tactile and kinesthetic sensations. If the olfactory and gustatory systems are connecting into the amygdala prenatally, I suggest that specifically somatosensory connections into the amygdala are also forming by the end of the first two months. Sufficient levels of tactile stimulation are provided by the maternal environment in the form of maternal contact comfort that releases early protoattachment behavior. Taylor (1987) notes, "The sensations impinging on the infant's skin presumably help regulate aspects of the infant's behavior and physiology" (p. 164). In accord with this, the classical work of Harlow (1958) demonstrates that skin-to-skin contacts come on-line early and that the infant actively seeks to adhere to as much skin surface on the mother's body as possible.

Most human fermales cradle their infants on the left side of the body (Manning et al., 1997; Harris, Almergi, & Kirsch, 2000). This tendency is well developed in women but not in men, is independent of handedness, and is widespread in all cultures. It has been suggested that this left-cradling tendency "facilitates the flow of affective information from the infant via the left ear and eye to the center for emotional decoding, that is, the right hemisphere of the mother" (p. 327). It also has been observed that "the language of mother and infant consist of signals produced by the autonomic, involuntary nervous system in both parties" (Basch, 1976, p. 766). This hemisphere, deeply connected into the ANS, is specialized for tactile perception on both sides of the body (Carmon & Benton, 1969) and for the perception and recall of spatial patterns of touch in nonverbal memory (Milner & Taylor, 1972). Again, the overt expressions of right hemisphere-to-right hemisphere communications are manifest from the very beginnings of infancy.

Neurobiological research indicates that "in early postnatal life, maintenance of critical levels of tactile input of specific quality and emotional content is important for normal brain maturation" (Martin, Spicer, Lewis, Gluck, & Cork, 1991, p. 3355). Indeed, the sensory input derived from contact with the mother during nursing has been suggested to shape dendritic growth (Greenough & Black, 1992). Infantile handling, tactile stimulation associated with comforting "holding" and "containing" experiences provided by the mother, induces permanent modifications of later hypothalamic CRF levels (Campbell, Zarrow, & Denenberg, 1973). Again, these experiences are right laterlized - Kalogeras et al. (1996) demonstrate that the right side of the human hypothalamus is dominant for neuropeptide secretion, including CRF activity.

I further propose that areas of the amygdala in the medial temporal lobe, especially the central and medial nuclei, are in a critical period of maturation that onsets in the last trimester of pregnancy and continues through the first two months of human life, the earliest period of bonding. In growth-facilitating perinatal environments, the experience-dependent maturation of interconnections between the infantis right amygdala and right paraventricular hypothalamic nuclei allow for co-regulation of vasopressin and the antistress hormone oxytocin in early maternal-infant interactions (Panksepp, 1998). This is a critical period of organization of the amygdalar-hypothalamic system, in which sensory information processed by the amygdala receives a positive or negative hedonic charge and is then relayed to various hypothalamic nuclei (Fonberg, 1986). These events occur in what Bowlby (1969) calls the "preattachment phase" of the first two months, the same time period of the evolution of Sternis (1985) "emerging self", and the time-frame of the first-to-mature homeostatic control system described by Brazelton (2000) that is on-line in the first weeks of life.

I previously cited a finding of a milestone for normal development in a rapid change in brain maturation at 8 weeks, reflecting the onset of a critical period during which synaptic connections in the primary visual cortex are modified by visual experience (Yamada et al., 1997, 2000). At this time infant visual preference behavior shifts from subcortical to cortical processing (Hoffmann, 1978), and face-to-face interactions, occurring within the primordial experiences of human play, first appear (Cohn & Tronick, 1987). Fogel and Branco observe

Three-month old-infants signal their willingness to engage in play by both gazing at mother and smiling, and they use gaze away and the cessation of smiling or the onset of crying to indicate their desire to end a bout of play. Before 3 months, infants do not have the ability to do this (1997, p. 76).

In these play sequences of affective synchrony, dyadically amplified elevations in sympathetic arousal occur in gaze engagements, followed by infant increases in parasympathetic activity (vagal nucleus ambiguus) in gaze aversion disengagements. The vagally-controlled hormone, oxytocin, now is released by "sensory stimuli such as tone of voice and facial expression conveying warmth and familiarity" (Uvnas-Molberg, 1997, p. 42). Vagal tone (Porges, 1991) is undeveloped and weak in the first quarter of the first year, but increases significantly at 2 to 4 months (Kagan, 1994), a time when primary intersubjectivity and delight and laughter first appear (Sroufe, 1996).

This same interval represents the onset of a critical period for the development of the anterior cingulate areas (see Figure 2) of the medial frontal cortex, a region involved in play and separation behaviors, laughing and crying vocalizations, face representations, and modulation of autonomic activity (MacLean, 1988, 1993; Paus, Petrides, Evans, & Meyer, 1993). MacLean (1987) provides evidence to show that this cortex is responsible for vocalizations that "maintain maternal-offspring contact." The anterior cingulate is also known to contribute to maternal behavior (Slotnick, 1967). A recent fMRI study reveals that the motheris cingulate and right orbitofrontal cortex respond to both pain and separation cries of an infant (Lorberbaum et al., 2000). With regard to the infantis expanding capacities, the right cingulate and parietal areas have been implicated in exploratory attentional movements (Gitelman et al., 1996) in the generation of a subjective prediction, and in the anticipation of being tickled (Carlsson et al., 2000).

Recall the earlier depiction of mutually regulated states of maternal-infant high arousal, attention, and vocalizations occuring in play experiences that emerge at this time (Feldman et al., 1999). This is also the onset of the positive resonances that occur within the mother-infant "protoconversations" that induce what Trevarthen calls primary intersubjectivity (Trevarthen, Aitken, Papoudi, & Roberts, 1998). In this relational context the primary caregiveris anterior cingulate-driven maternal behavior would be socially tuning the infantis medial frontal cortex, thereby influencing the parcellation and final circuit wiring of the babyis developing anterior cingulate. During this critical period of the onset the infantis anterior cingulate-right temporal (Nakamura et al., 2000) face processing, which Mahler et al., call the symbiotic period, the infant forms a discriminate attachment to the motheris face.

The later occurring parasympathetic phase of the critical period of growth of this limbic component would occur in the third quarter, a time of cingulate-driven expressions of separation-anxiety (MacLean, 1990; Joseph, 1992), responses to attachment ruptures. At 7 to 10 months infants show fear (Sroufe, 1996) and stranger anxiety, in which they inhibit ongoing behavior and withdraw when exposed to novel and threatening situations and unfamiliar people. The emergence of this more complex defensive behavior, inhibited approach, represents the parasympathetic maturation of the cingulate.

Furthermore, in light of the known role of the cingulate in consciousness (Kennard, 1955) it is tempting to speculate that the experience-dependent maturation of this limbic structure may be activated in moments of dyadically expanded states of consciousness that onset in the middle of the first year. Tronick et al. (1998) are now describing how microregulatory social-emotional processes of communication literally expand intersubjective states of consciousness in the infant-mother dyad. They argue that the babyis self-organizing system, when coupled with the motheris, allows for a brain organization which can be expanded into more coherent and complex states of consciousness. The interpersonal context of a co-regulated dyadic system allows for "a mutual mapping of (some of) the elements of each partneris state of consciousness into the other partneris brain" (p. 296).

I suggest that Tronick is describing an expansion of what Edelman (1989) calls primary consciousness. Edelman states primary consciousness relates visceral and emotional information pertaining to the biological self to stored information pertaining to outside reality, and that it is lateralized to the right brain. Activity of limbic cingulate areas are known to be associated with primary consciousness (Denton et al., 1999). This developmental work supports the idea that consciousness is a product of that part of the brain that handles human relations, and is a property of a brain that is and has been in communication with other brains (Barlow, 1980; Schore, 1994). It also suggests that indices of the maturation of the infantis capacity for primary consciousness needs to be included in our models of infant mental health.

The critical period of anterior cingulate-driven limbic maturation thus overlaps Bowlbyis (1969) phase of "attachment-in-the-making", and mediates what Stern (1985) terms, "the core self." Brazelton (2000) describes the emergence in the second quarter of the first year of a second homeostatic control system, one associated with a mutual reciprocal feedback system. Though an advance of the former control system it is still "an immature psychophysiological system." I suggest this system can be identified as a maturing anterior cingulate which now hierarchically controls the earlier amygdala-dominated limbic configuration.

The right insula, a limbic structure involved in emotional and facial processing (Berthier, Starkstein, & Leiguarda, 1987), in integrating tonal structure with a speakeris emotions and attitudes (Riecker et al., 2000), and in visceral and autonomic functions that mediate the generation of an image of oneis physical state (Craig et al., 2000) is also activated in primary consciousness (Denton et al., 1999). This limbic structure is implicated in pain processing and serves as an alarm center, "alerting the individual to potentially distressing interoceptive stimuli, investing them with negative emotional significance" (Banzett at al., 2000, p. 2120). It is tempting to speculate that the experience-dependent maturation of this system is associated with both the more complex representation of body image and "stranger anxiety" that emerges in the second half year of life.

In the last quarter of the first year the quality of the infantis social relatedness changes dramatically (see Schore, 1994), due to the concurrent rapid myelination and maturation of developing limbic and cortical association areas. If earlier face-to-face interactions contain only spontaneous communication processes, after 9 months the infant can engage in "joint attention", the ability to shift attention between an object and a person. In this form of nonverbal communication the infant coordinates his visual attention with that of the caregiver, and is now not only aware of an object but simultaneously aware of the motheris attention to the object. In such instances of what Trevarthen et al. (1998) call "secondary intersubjectivity", each member of the dyad co-aligns separable, yet related forms of consciousness.

Joint attention occurs within highly affectively-charged social referencing transactions, an attachment process that mediates a resonance of positive affect (Schore, 1994). This dyadic mechanism allows the infant to appreciate that "the other person is a locus of psychological attitudes toward the world, that the other is eattendingi in such a way that shared experiences are possible" (Hobson, 1993, p. 267). In this manner the child comes to understand others "as intentional beings, that is, as subjects of experience possessing internal states such as interest and attention" (Tomasello & Camaioni, 1997, p. 20). In order to get an adult to tune into his attentional and intentional focus on the world the infant now uses an expanded repertoire of bidirectional communicative gestures, an important cognitive advance that communicates intention (Goldin-Meadow, 2000). It is also is in this period, the last quarter of the first year, when "the infant starts to adopt a mentalistic strategy to interpret and predict the behavior of other agents" and is "capable of taking the intentional stance" (Gergely et al., 1995, p. 184). In developmental neurobiological research, Caplan et al. suggest that "the development of joint attention might reflect maturation of the prefrontal cortex" (1993, p. 589).

These critical advances represent a further maturation of the right hemisphere, since current research suggests it contributes to attention and intention (Mattingley, 1999; Sturm et al., 1999). Very recent studies of joint attention demonstrate that the right (and not left) hemisphere shifts attention to where someone is looking to follow the gaze of another (Kingstone, Friesen, + Gazzaniga, 2000). In fact there is now evidence for "a special role for the right frontal lobe in sustaining attention over time" (Rueckert & Grafman, 1996, p. 952). Very recent studies now reveal that the right orbitofrontal and right anterior insula cortices are components of a neural circuit that "enables integration of adaptive bodily responses with ongoing emotional and attentional states of the organism" (Critchley et al., 2000b, p. 3033). But in addition this right prefrontal cortex is fundamentally involved in "regulating emotional responses" (Hariri et al., 2000).

The Maturation of an Orbitofrontal Regulatory System

In Affect Regulation and the Origin of the Self and continuing works I offer evidence to show that the orbital prefrontal cortex enters a critical period of growth which spans the last quarter of the first through the middle of the second year, an interval that corresponds with the beginnings of human socialization. The critical period of orbitofrontal-driven limbic maturation thus overlaps and mediates what Stern (1985) terms the developmental achievement of "the subjective self." This prefrontal limbic structure is reciprocally interconnected with other limbic areas in the insula (Augustine, 1996), anterior cingulate (Devinsky et al., 1995), and the amygdala (Barbas & de Olmos, 1990), and represents the hierarchical apex of the limbic system.

Brothers (1995, 1997) describes a limbic circuit of orbitofrontal cortex, anterior cingulate gyrus, amygdala, and temporal pole which functions as a social "editor" that is "specialized for processing othersi social intentions" by appraising "significant gestures and expressions" (Brothers, 1997, p. 27) and "encourages the rest of the brain to report on features of the social environment" (p. 15). The editor acts as a unitary system "specialized for responding to social signals of all kinds, a system that would ultimately construct representations of the mind" (p. 27). Mesulam points out that the prefrontal areas involved in emotional modulation and attentional functions help to create "a highly edited subjective version of the world" (1998, p. 1013). This emergent subjective function is the outcome of a secure attachment.

In a recent entire issue of Cerebral Cortex on "The mysterious orbitofrontal cortex," the editors conclude that "the orbitofrontal cortex is involved in critical human functions, such as social adjustment and the control of mood, drive and responsibility, traits that are crucial in defining the epersonalityi of an individual" (Cavada & Schultz, 2000, p. 205). Referring back to Brazelton and Crameris conce