by Bruce D. Perry, M.D., Ph.D.
Ronnie A. Pollard, M.D.
Toi L. Blakley, M.D.
William L. Baker, MS
ChildTrauma Academy Programs
Department of Psychiatry and Behavioral Sciences
Baylor College of Medicine
Texas Children's Hospital
Houston Veteran's Affairs Medical Center
07/25/95, In Press: Infant Mental Health Journal, 1996. Note that this online version may have minor differences from the published version.
Bruce D. Perry, M.D., Ph.D.
Department of Psychiatry and Behavioral Sciences
Baylor College of Medicine
One Baylor Plaza
Houston, TX 77030
(713) 770-3747 FAX
Childhood trauma has profound impact on the emotional, behavioral, cognitive, social and physical functioning of children. Developmental experiences determine the organizational and functional status of the mature brain. The impact of traumatic experiences on the development and function of the brain are discussed in context of basic principles of neurodevelopment. There are various adaptive mental and physical responses to trauma, including physiological hyperarousal and dissociation. Because the developing brain organizes and internalizes new information in a use-dependent fashion, the more a child is in a state of hyperarousal or dissociation, the more likely they are to have neuropsychiatric symptoms following trauma. The acute adaptive states, when they persist, can become maladaptive traits. The clinical implications of this new neurodevelopmental conceptualization of childhood trauma are discussed.
Adults interpret the actions, words and expressions of children through the distorting filter of their own beliefs. In the lives of most infants and children these common adult misinterpretations are relatively benign. In many cases, however, these misinterpretations can be destructive. The most dramatic example occurs when the impact of traumatic events on infants and young children is minimized. It is an ultimate irony that at the time when the human is most vulnerable to the effects of trauma -- during infancy and childhood -- adults generally presume the most resilience.
This destructive misperception has permeated the mental health field. In the last ten years, our society has spent billions of dollars studying and treating adult trauma victims, primarily male combat veterans -- this despite the fact that many more females are traumatized by rape in our society than males by combat. In comparison, few resources have been dedicated to research or treatment focusing on childhood trauma, and only a fraction of those on studying or treating the traumatized infant (Perry, 1994a; Perry, 1995).
The purpose of this paper is to discuss various aspects of the impact of traumatic experiences on infants and young children focusing specificically on the relationships between neurodevelopment and traumatic experience. The conceptual views articulated in this paper represent an evolving understanding based upon our extensive clinical experience with young children and infants who have been severely traumatized (n = 175). To some extent, these conceptual views are based upon well-established principles of neurodevelopment applied within the context of less clearly delineated ideas (and data) emerging in the field of traumatology. This is intended as only a preliminary perspective to guide future studies in the clinical phenomenology and neurobiology of child maltreatment.
Millions of children across the world are exposed to traumatic experiences. These may be pervasive and chronic (e.g., course-of conduct maltreatment such as incest; war) or timelimited (e.g., natural disaster, drive-by shooting). Conservative estimates of the number of children in the United States exposed to a traumatic event in one year exceed 4 million (Perry, 1994a). These experiences -- physical or sexual abuse, living in the fallout zone of domestic or community violence, surviving a serious car accident -- all have an impact on the child's development (Taylor, Zuckerman, Harik, and Groves, 1992; Pynoos, Frederick, Nader, Arroyo, Steinberg, Eth, Nunez, and Fairbanks, 1987; Osofsky, 1995). Depending on the severity, frequency, nature, and pattern of traumatic events, at least half of all children exposed may be expected to develop significant neuropsychiatric symptomatology (Schwarz and Perry, 1994) Children exposed to sudden, unexpected man-made violence appear to be more vulnerable -making the millions of children growing up with domestic- violence or community violence at great risk for profound emotional, behavioral, physiological, cognitive and social problems.
One of the most studied neuropsychiatric syndromes which develops following trauma is post-traumatic stress disorder (PTSD). The majority of this work has been with adult combat veterans (Krystal, Kosten, Perry, Mason, Southwick, and Giller, 1989; DaCosta, 1871). In the last five years, however, childhood PTSD has been widely observed in various popuiations of traumatized or maltreated children (McFarlane, 1987). Children exposed to trauma may have a range of PTSD symptoms, behavior disorders, anxieties, phobias, and depressive disorders (Schwarz and Perry, 1994). This includes children who were kidnapped (Terr, 1983), witnessed violent crime (Schwarz and Kowalski, 1991), have been abused (Browne and Finkelhor, 1986; Kiser, Heston, and Millsap, 1991) or survived other severe trauma (Kaufman, 1991). Traumatic experiences in childhood increases the risk of developing a variety of neuropsychiatric symptoms in adolescence and adulthood (Davidson and Smith, 1990; Famularo, Kinscherff, and Fenton, 1991; Ogata, Silk, Goodrich, Lohr, Westen, and Hill, 1990; Teicher, Glod, Surrey, And Swett, 1993).
Trauma is an experience. How is it that this experience can transform a child's world into a terror-filled, confusing miasma that so dramatically alters the child's trajectory into and throughout adult life? Ultimately, it is the human brain that processes and internalizes traumatic (and therapeutic) experiences. It is the brain that mediates all emotional, cognitive, behavioral, social and physiological functioning. It is the human brain from which the human mind arises and within that mind resides our humanity. Understanding the organization, function and development of the human brain, and brain-mediated responses to threat are the keys to understanding the traumatized child.
The human brain is an amazingly complex organ comprised of over 100 billion neurons and ten times as many goat cells, all organized into systems designed to sense, process, store,perceive, and act on information from the external (e.g., visual,-tactile, olfactory, auditory) and the internal (e.g., hormonal signals associated with hunger) environment (see Appendix 1, Key Points: Brain Organization and Function). All of these complex systems and activities work together with one overarching purpose -- survival (Goldstein, 1995). The major working units of the brain are neurons, and neurons are interconnected into networks, and networks into systems, and systems work together to mediate a set of specific functions (e.g., vision). Different systems in the brain mediate different functions. Systems in the frontal cortex are involved in abstract thought; systems in the brainstem are responsible for regulating heartrate, blood pressure and arousal states; systems in the limbic areas are responsible for attachment, affect regulation and aspects of emotion; and systems in the cortex are responsible for abstract cognition and complex language. Each of these systems vary somewhat with regard to function, specific primary neurotransmitter networks, synaptic structure, and regional localization. All of these neural system, however, function in a similar fashion on the molecular level. They obey similar molecular rules mediating development, changes in response to chemical signals, and storage of information (see Table 1). The single most significant distinguishing feature of all nervous tissue -- of neurons -- is that they are designed to change in response to external signals. Those molecular changes permit the storage of information by neurons and neural systems. Indeed it is this capacity which allows the brain to be responsive to the environment (external and internal) to allow survival of the organism.
All experience is filtered by our senses. All sensory signals (e.g., sound, sight, taste, touch), in turn, initiate a cascade of cellular and molecular processes in the brain that alter neuronal neurochemistry, cytoarchitecture and, ultimately, brain structure and function. This process of creating some internal representation of the external world (i.e., information) depends upon the pattern, intensity and frequency of neuronal activity produced bysensing, processing and storing signals. The more frequently a certain pattern of neural activation occurs, the more indelible the internal representation. Experience thus creates a processing template through which all new input is filtered. The more a neural network is activated, the more there will be use-dependent internalization of new information needed to promote survival (Cragg, 1975).
The most common examples of this use-dependent storage of information are learning and cognitive memory. A related phenomenon is sensitization. A sensitized neural response results from a specific pattern of repetitive neural activation or experience. Sensitization occurs when this pattern of activation results in an altered, more sensitive system. Once sensitized, the same neural activation can be elicited by decreasingly intense external stimuli (Kalivas and Kuffy, 1989; Kleven, Perry, Woolverton, and Seiden, 1990). In a very real sense, traumatized children exhibit profound sensitization of the neural response patterns associated with their traumatic experiences. The result is that full-blown response patterns (e.g., hyperarousal or dissociation) can be elicited by apparently minor stressors.
Sensitization may result when experience activates neurosensory apparatus, altering the pattern and quantity of neurotransmitter release throughout neuronal systems responsible for sensation, perception, and processing of that specific experience. Trauma will produce such a result. Neurotransmitter receptor/effector activation then alters intracellular chemical constituents, including second (e.g. cAMP, phosphatidyl inositol) and third messengers (e.g., calcium). Changes in these messengers alter the micro-environment of the nucleus, inducing alterations in gene transcription and expression of proteins involved in both neuronal structure and functioning. This, in turn, may cause sensitization of neurotransmitter receptor/effectors to similar future neurotransmitter stimulation in all interconnected neural systems (LeDoux, Cicchetti, Xagoraris, and Romanski, 1989; LeDoux, Romanski, and Xagoraris, 1990).
Use-dependent internalization of the fear response -- a 'state' memory -- can be built into a mature brain (e.g., combat-related PTSD). In thedeveloping brain, these states organize neural systems, resulting in traits. The human brain exists in its mature form only as a byproduct of genetic potential and environmental history. Experiences, including traumatic, which may cause sensitization or learning in the mature brain will, during development, determine functional capacity of the human brain. In essence, the same unique molecular characteristics of nervous tissue which allow the mature brain to store new information are those responsible for organizing the brain during development (Goelet and Kandel, 1986; Kandel and Schwartz, 1982).
In the developing brain, undifferentiated neural systems are critically dependent upon sets of environmental and micro-environmental cues (e.g., neurotransmitters, cellular adhesion molecules, neurohormones, amino acids, ions) to appropriately organize from their undifferentiated, immature forms Appendix 1, Key Points: Brain Development). Lack (or disruption) of these critical cues can result in abnormal neuronal division, migration, differentiation, synaptogenesis -- all of which contribute to malorganization and compromised function in the affected systems (Cragg, 1975; Lauder, 1988; Perry, Wainwright, Won, Hoffman and Heller, 1990b). Two major principles of neurodevelopment related to the timing and nature of these organizing cues are 1) use-dependent development and organization of the brain and 2) critical and sensitive periods.
The brain develops in a sequential and hierarchical fashion -- i.e., from less complex (brainstem) to most complex (limbic, cortical areas). These different areas develop, organize and become fully functional at different times during childhood. At birth, for example, the brainstem areas responsible for regulating cardiovascular and respiratory function must be intact for the infant to survive, and any malfunction is immediately observable. In contrast, the cortical areas responsible for abstract cognition have years before they will be 'needed' or fullyfunctional. This means that there are different times during which different areas of the CNS are organizing and, therefore, either require (critical periods) or are most sensitive to (sensitive periods) organizing experiences (and the neurotrophic cues related to these experiences). Disruptions of experience-dependent neurochemical signals during these periods may lead to major abnormalities or deficits in neurodevelopment -- some of which may not be reversible (see below). Disruption of critical cues can result from 1) lack of sensory experience during critical periods or, more commonly, 2) atypical or abnormal patterns of neuronal activation due to extremes of experience (e.g., child maltreatment).
The simple and unavoidable result of this sequential neurodevelopment is that the organizing, sensitive brain of an infant or young children is more malleable to experience than a mature brain. While experience may alter the behavior of an adult, experience literally provides the organizing framework for an infant and child. Because the brain is most plastic (receptive to environmental input) in early childhood, the child is most vulnerable to variance of experience during this time.
Deprivation of critical experiences during development may be the most destructive yet the least understood area of child maltreatment. Unlike broken bones, irreversible maldevelopment of brain areas mediating empathy resulting from emotional neglect in infancy and childhood is not readily observable. Ironically, while rarely studied in humans, the neurodevelopmental impact of extremes of sensory deprivation is the subject of hundreds of animal studies. These studies include disruptions of visual stimuli (Coleman and Riesen, 1968), environmental enrichment (Altman and Das, 1964; Cummins and Livesey, 1979), touch (Ebinger, 1974), and other factors (Sapolsky, Krey, and McEwen, 1984; Sapolsky, Uno, Rebert, Finch, 1990). These studies illustrated that a very narrow window - a critical period - exists during which specific sensory experience was required for optimal organization and development of the part of the brain mediating a specific function. While these phenomenon have been examined in great detail for the primary sensory modalities in animals, similar usedependent neurodevelopment occurs in all parts of the human central nervous system. Abnormal micro-environmental cues and atypical patterns of neural activity during critical and sensitive periods, then, can result in malorganization and compromised function in brainmediated functions such as humor, empathy, attachment and affect regulation .
Some of the most powerful clinical examples of this phenomemon are related to lack of attachment experiences early in life. The child who has been emotionally neglected early in life will exhibit profound attachment problems which are extremely insensitive to any replacement experiences later in life, including therapy. Examples of this include feral children, children in orphanages observed by Spitz and Wolf (1946) and, often, the remorseless, violent child (Perry, Blakley, Pollard, in press).
While deprivations and lack of specific sensory experiences are common in the maltreated child, the traumatized child experiences developmental insults related to discrete patterns of over-activation of neurochemical cues. Rather than adeprivation of sensory stimuli, the traumatized child experiences over-activation of important neural systems during sensitive periods of development.
The human body and human mind have sets of very primitive, deeply ingrained physical and mental responses to threat (Appendix 1: Key Points: The Response to Trauma). These physiological and mental reactions to danger have been best characterized in adult humans or animal models. The most familiar set of responses to threat has been labeled the fight or flight' reaction. Indeed, despite a great deal of animal and descriptive clinical data in humans illustrating the heterogeneity of the response to stress (Mason, 1971; Goldstein, 1995), an inordinate clinical focus has been placed on the 'fight or flight' response -- a pattern commonly seen in adult, male mammals (Cannon, 1914; Selye, 1936).
There are, of course, other response-sets to threat. Indeed, infants and children are much less likely to use a classic 'fight or flight' response -- it is not very practical. At different stages of development, and in the face of different stressors, response patterns will vary. Two major neuronal response patterns important for the traumatized child, the hyperarousal continuum and the dissociative continuum, are described below.
In the initial stages of threat, an alarm reaction is initiated (Table 1). The alarm reaction is characterized by a large increase in activity of the sympathetic nervous system, resulting in increased heart rate, blood pressure, respiration, a release of stored sugar, an increase in muscle tone, a sense of hypervigilance, and tuning out of all non-critical information. All of these actions prepare the body for defense -- to fight with or run away from the percieved or sensed threat. If the threat materializes, a full fight or flight response may be activated.
This complex set of interactive processes includes activation of the centrally-controlled peripheral autonomic nervous system, the immune system (Giller, Perry, Southwick, and Mason, 1990), the hypothalamic-pituitary axis with a concomitant peripheral release of adrenocorticotrophic hormone and cortisol (Sapolsky et al., 1984), and other stress-response neural systems in the brain (Krystal, Kosten, Perry, Mason, Southwick, Giller, 1989; Perry, Southwick, and Giller, 1990a). The locus coeruleus (LC) is the key mediator of the classic"fight or flight" response to threat (Svensson, 1987; Korf, 1976; Elam, Svensson, Thoren, 1984; Altman, Das, 1964). This bilateral grouping of norepinephrine-containing neurons originates in the pons, a more primitive, regulatory part of the brain, and sends diverse axonal projections to virtually all major brain regions, enabling its function as a general regulator of noradrenergic tone and activity (Moore and Bloom, 1979). The ventral tegmental nucleus (VTN) also plays a role in regulating the sympathetic nuclei in the pons/medulla.
The 'sensitized' hyperarousal response: If a child faced with threat responds with a hyperarousal response, there will be a dramatic increase in LC and VTN activity. This increased release of norepinephrine regulates the total body response to the threat. The brain regions involved in the threat-induced hyperarousal response play a critical role in regulating arousal, vigilance, affect, behavioral irritability, locomotion, attention, the response to stress, sleep, and the startle response (Andrade and Aghajanian, 1984; Bhaskaran and Freed, 1988). Initially following the acute fear response, these systems in the brain will bereactivated when the chiid is exposed to a specific reminder of the traumatic event (e.g., gunshots, the presence of a past perpetrator). Furthermore, these parts of the brain may bereactivated when the child simply thinks about or dreams about the event. Over time, these specific reminders may generalize (e.g., gunshots to loud noises, a specific perpetrator to any strange male). In other words, despite being distanced from threat and the original trauma, the stress-response apparatus of the child's brain is activated again and again.
This use-dependent activation of these areas leads to sensitization, and sensitization of catecholamine (LCNTN-amygdaloid) systems leads to a cascade of associated functional changes in brain-related functions (Vantini, Perry, Gucchait, U'Prichard, and Stolk, 1984; Perry, Stolk, Vantini, Gucchait, and U'Prichard, 1983). This sensitization of the brain stem and midbrain neurotransmitter systems also means the other critical physiological, cognitive, emotional and behavioral functions which are mediated by these systems will become sensitized. As brain areas involved in the acute stress response also mediate a variety of other functions, sensitization of these systems by repetitive re-experiencing of a traumatic event leads to dysregulation of these functions. A traumatized child may, over time, exhibit motor hyperactivity, anxiety, behavioral impulsivity, sleep problems, tachycardia, hypertension and a variety of neuroendocrine abnormalities (Perry, 1994a; Perry and Pate, 1994b; DeBellis, Lefter, Trickett, Putnam, 1994; DeBellis, Chrousos, Dorn, Burke, Helmers, Kling, Trickett, Putnam, 1994; Ito, Teicher, Glod, Harper, Magnus, Gelbard, 1993; Hoffman, DiPiro, Tackett, Arrendale, and Hahn, 1989).
This also means, of course, that these components of the fear response, themselves, become sensitized. Everyday stressors which previously may not have elicited any response now elicit an exaggerated reactivity -- these children are hyperreactive and overly sensitive. This is due to the fact that, simply stated, the child is in apersisting fear state (which is now a 'trait'). Furthermore, this means that the child will very easily be moved from being mildly anxious to feeling threatened to being terrorized. In the long run, what is observed in these children is a set of maladaptive emotional, behavioral and cognitive problems which are rooted in the original adaptive response to a traumatic event.
Abnormal persistence of the original hyperarousal seen during the trauma can explain some, but not all of the symptoms traumatized children develop. indeed, this can not explain the majority of neuropsychiatric problems which result from pervasive trauma in infancy. Why do more maltreated boys than maltreated girls get referred to the mental health system; why do more males exhibit evidence of sensitized hyperarousal systems (motor hyperactivity, behavioral impulsivity, hypervigilance) following trauma; why do more females exhibit evidence of sensitized dissociative systems (avoidant, depressive, dissociative) following trauma ? The answers to these questions are likely to be rooted in understanding the other adaptive response patterns seen in the face of trauma -- patterns that may be more adaptive for children or females than for adult males.
This total body, neurophysiological hyperarousal response exists because it promotes survival. Clearly, if you are an adult male human, it is highly adaptive, in the face of threat, to mobilize a total body response to fight or flee. It is not very adaptive, however, if you are a child or even, as humans evolved, a female. It has been our experience that the hyperarousal response - - with the primary adaptive pattern being the defenses of 'fight or flight' -- is not the primary response to threat in young children and infants. Indeed, young children more commonly utilize a combination of adaptive responses which are designed to, in the early stages of threat, bring caretakers to defend them (an initial hyperarousal response). And, if the threat continues, to move through a dissociative continuum, initially becoming immobile (freezing) and compliant, later completely dissociated; finally, in the extreme, fainting. These responses have been best characterized in animals and have been labeled the defeat or giving up response (Miczek, Thompson, and Tornatzky, 1990). In humans, we will call this set of behaviors a 'surrender' response.
Children, of course, are not particularly well equipped to fight or flee. In the initial stages of distress, a young child will use vocalization, i.e., crying, to get a caretaker to know that they are under threat. This is a successful adaptive response to threat if the caretaker comes and either fights for, or flees with, the young threatened child. Crying, therefore, is a developmentally appropriate response to a threat which the child is unable to avoid. Unfortunately, crying infrequently will bring an adult to defend a traumatized child. Indeed, for many of the maltreated children we work, crying for"help" from a potential trauma is doomed to fail -- often the parent causes the trauma. In the absence of an appropriate caretaker response to their initial alarm outcry, the child, eventually after many painful disappointments, will abandon this behavior (a defeat or surrender response). In the face of persisting threat and, depending upon the age of the child and the nature of the threat, the child will move along the hyperarousal continuum (the child's version of 'fight or flight') or into the dissociative continuum (Figure 1).
Freezing and oppositional-defiant behaviors: A first reaction in the face of continuing threat may be to freeze. The adaptive advantage of this is clear. Freezing allows better sound localization, keener visual observation -- and environmental scan for potential threat. In addition, lack of movement is a form of camouflage, reducing the chance of attracting a predator. In the face of escalating threat, increasing anxiety and decreasing cognitive processing, the freeze response can confer adaptive advantage by allowing one to 'organize' and 'figure out' how to respond.
Children who have been traumatized and have developed a 'sensitized' hyperarousal or 'sensitized' dissociative pattern will often use this freezing mechanism when they feel anxious. This is often labeled oppositional-defiant behavior. The child will feel anxious due to an evocative stimulus to which their sensitized neural systems are reacting (e.g., a family visit). They are often not aware of the evocative nature of a given event, but what they do experience - deeply -- is anxiety. At this point, they tend to feel somewhat out of control and will cognitively (and often, physically) freeze. When adults around them ask them to comply with some directive, they may act as if they haven't heard or they"refuse". This forces the adult -- a teacher, a parent, a counselor -- to give the child another set of directives. Typically, these directives involve more threat. The adult will say, "If you don't do this, l will...". The nonverbal and verbal character of this 'threat' make the child feel more anxious, threatened and out of control. The more anxious the child feels, the quicker the child will move from anxious to threatened, and from threatented to terrorized (see Table 1). If suffficiently terrorized, the 'freezing' may escalate into complete dissociation.
Dissociation: Among the constellation of symptoms associated with the trauma response in adults is dissociation. Dissociation is simply disengaging from stimulit in the external world and attending to an 'internal' world. Daydreaming, fantasy, depersonalization, derealization, and fugue states are all examples of dissociation (Putnam, 1991). There are gradations of dissociation -- from simple daydreaming to profound torture-induced loss of conciousness. Common examples of dissociation in the face of threat have been described following combat. A soldier in the midst of a fire fight may-be able-to engage in combat and only after the event is over, will he feel his heart racing and look down and see the wound in his leg. Soldiers often utilize an automated, automatic, detached response set. It is during those moments, literally minutes of combat, that the soldier will tell you that there was sound going on all around him, but it seemed distant, the event took place as if on a video screen. It is this very ability to dissociate (and utilize a previously internalized set of cognitive and behavioral responses, drilled in to them) which can keep soldiers alive.
The capacity to dissociate in the midst of terror appears to be a differentially available adaptive response. Some people dissociate early in the arousal continuum -- some people dissociate only in the state of complete terror. Viewed dispassionately, because of the diminished cognitive capacity of an adult in the full-blown fight or flight response, there is great teleological logic in a partial dissociative response (it is what allows the soldier to fight without panic).
It has been our experience with infants and young children that the behaviors exhibited in the acute and post-acute trauma include numbing, compliance, avoidance, and restricted affect, all consistent with a primary dissociative response pattern. Traumatized children use a variety of dissociative techniques. Children report going to a 'different place,' assuming persona of heroes or animals, a sense of 'watching a movie that I was in' or 'just floating' - ciassic depersonalization and derealization responses. Observers will report these children as numb, robotic, non- reactive, "day dreaming", "acting like he was not there","staring off into space with a glazed look." If immobilization, inescapability or pain are involved, the dissociative responses will become more predominant.
Neurobiology of the dissociative continuum: The neurobiology of the hyperarousal response involves the catecholamines originating in the brainstem (Murberg, McFall, and Veith, 1990). The neurobiology of the dissociative continuum is different. In animals, the 'defeat' response has a distinct neurobiology which appears to match what little is known about the comparable dissociation response in humans. Indeed, the neurobiology and phenomenology of dissociation appears to most approximate the 'defeat' reaction described in animals (Henry, Stephens, and Ely, 1986; Heinsbroek, van Haaren, Feenstra, and Boon, 1991; Miczek et al., 1990).
As with the hyperarousal/fight or flight response, dissociation involves brainstemmediated CNS activation which results in increases in circulating epinephrine and associated stress steroids (Glavin, 1985; Henry, Liu, Nadra, Qian, Mormede, Lemaire, Ely, and Hendley, 1993; Herman, Guillonneau, Dantzer, Scatton, Semerdjian-Rouquier, and Le Moal., 1982). A major CNS difference, however, is that, in dissociation, vagal tone increases dramatically, decreasing blood pressure and heart rate (occasionally resulting in fainting) despite increases in circulating epinephrine. In addition, there appears to be an increased relative importance of dopaminergic systems, primarily mesolimbic and mesocortical (Kalivas, 1985; Kalivas, Richardson-Carlson, and Van Orden, 1986; Kalivas, Duffy, Dilts and Abhold, 1988; Abercrombie, Keefe, DiFrischia, and Zigmond 1989). These dopaminergic systems are intimately involved in the reward systems, affect modulation (e.g., cocaine-induced euphoria) and, in some cases, are co-localized with endogenous opioids mediating pain and other sensory processing. These opioid systems are clearly invoived in altering perception of: painful stimuli, sense of time, place and reality. Indeed, most opiate agonists can induce dissociative responses. Of primary importance in mediating the freeze or surrender dissociative response are these endogenous opioid systems (Abercrombie and Jacobs, 1988). There are also a variety of other brainstem, midbrain and limbic region neurotransmitters which must be involved in this complex set of responses. The important point to keep in mind is that whatever the neurobiology of these dissociative continuum responses are, they are distinct from the neurobiology of hyperarousal.
Why do different individuals use distinct response patterns in the face of threat ? Age seems to play a role. Our clinical experience suggests that the younger an individual is, the more likely they are to use dissociative adaptations over hyperarousal responses. The nature of the trauma seems to be important to the pattern of adaptation; the more immobile, helpless, and powerless the individual feels, the more likely they are to utilize dissociative responses. When physical injury, pain or torture (hence, opioid activation) is involved in the traumatic experience, an individual will be more likely to use dissociative responses. Finally there is a clear sex difference in response patterns; females utilize dissociative adaptations more than males. Some insight into these clinical observations can be found in examining the relationship between these responses and the underlying purpose of all brain related functions -- survival. In order to persist over thousands of generations, each response pattern must have some adaptive advantages.
It is easy to see the adaptive advantage conferred by the aggressively defensive hyperarousal/fight or flight response in adult males. One can only imagine what would have happened to the human species in the face of threat if adult males always dissociated in the face of threat. A group of numb, passive and immobile humans would be easy prey for natural predators.
Humans evolved over the last 250,000 years in the presence of two major predators: large cats (e.g., tigers, panthers) and, the most dangerous predators, other hominids, including humans (see Leakey, 1994). To the cats, all humans (males, females and children) -- were roughly equivalent prey, with some preferences for the small, slow and weak. To other hominids, however, there was a dramatic difference between males, females and young children. As described extensively in anthropological literature, it was likely a common practice for clans of hominids to raid a competing clan's camp, drive away or kill the males and take the females and young children as property (not unlike the recent history of Western 'Civilization').
It promoted survival of the species if young children and females survived these raids. It was more adaptive for children to dissociate and surrender than to be hyperaroused and try a fight or flight response. In the face of threat, it was self-protective to become numb, nonhysterical, compliant, obedient and not combative. Running would result in isolation and sure death. Fighting would be futile.
The same is likely true for adult females. One need not imagine long the response of a violent human male when faced with one female who will willingly comply with the commands to move to his camp and with another who is screaming, yelling, hitting, fighting and trying to run away. Hyperarousal (fighting or fleeing) would clearly reduce the 'property' value of a female, reducing the probability that her genes would be passed to another generation. Both the hyperarousal and the dissociative continuum were selected as adaptive advantages through thousands of generations of clan/tribal intra-species warfare.
As the human animal matures and grows more capable of fleeing and fighting, the predominance of the dissociative adaptation appears to diminish. It continues to be an extremely important adaptive response in the face of threat, however. In large part because dissociation allows one to maintain or even diminish the internal state of physiological hyperarousal, thereby allowing cognition and problem-solving at a higher level of capability than would be possible in a state of absolute terror.
While there are undoubtedly many other possible reasons for the evolution of dissociation and hyperarousal response, there are clear sex and age differences in predominant adaptive style. This is obvious when examining epidemiological data related to the incidence of neuropsychiatric disorders in male compared to female children. The three to one (male to female) ratio of childhood neuropsychiatric problems disappears in adolescence and, suddenly, in early adulthood shifts to become two to one (female to male). In childhood more boys meet diagnostic criteria for externalizing disorders such as ADHD, conduct disorder, and oppositional defiant disorder while more girls have a higher incidence of internalizing disorders such as depressive, anxiety or dissociative disorders.
The vast majority of young children from backgrounds of abuse and neglectand other trauma who present to the mental health system with symptoms of aggression, inattentiveness and noncompliance are male. They typically are diagnosed with attention deficit hyperactivity disorder (ADHD). One wonders what happens to all the young girls who have been similarly traumatized. Children present to the mental health system because some adults in their world have been upset by their symptoms (which have almost always been caused by other adults). A compliant, dissociative, depressed young girl will generallynG be brought to the attention of the mental health system, while her combative, verbally abusive and behaviorally-impulsive hyperaroused sibling (coming from the exact same abusive setting) will be. The potential homicide threatens, the potential suicide inconveniences.
We have followed approximately 50 young children in the acute post-traumatic period. There is a clear relationship between the type of adaptive response utilized in the acute situation and the symptoms which persist at six months (see Figure 2); Perry, 1994c; 1995). In our studies utilizing heart rate as a non-invasive indicator of physiological hyperarousal, we have seen that as a child begins to dissociate their heart rate plateaus and in many cases drops. As the dissociating child decreases their focus on external threatening cues and increasingly attends to internal cues (e.g., their special place), they feel less threatened and effectively alter the neurobiological pattern of response to threat, moving along the dissociative continuum from freezing to surrender. This is in sharp contrast to children using a predominant hyperarousal response. These children's heart rates are elevated at baseline and are extremely reactive, dramatically increasing in the face of emotionally-laden cognitive or physical cues (Perry, 1994c; 1995) As these children move along the hyperarousal continuum, from alarm-vigilance to fear and terror, they continue to use the neurobiological pattern of the well-described 'fight or flight' response.
Because the neurobiology and response patterns of dissociation are distinct from those associated with the hyperarousal continuum, one would expect different sets of symptoms to result from the use-dependent internalization of any persisting response to threat. This is indeed the case. The Diagnostic and Statistical Manual of Mental Disorders IV (DSM IV) labels routinely applied to the population of severely maltreated children with whom we work vary dramatically. Clearly the DSM IV does not have adequate descriptive categories for the majority of trauma- related neuropsychiatric syndromes observed in children, especially from deliberatelyinflicted trauma. Only 70 percent of our severely traumatized children (all ages) with dramatic symptoms of physiological hyperarousal (n=120; males/females: 4:1) meet diagnostic criteria for PTSD (Perry, 1995). Furthermore, in our experience, females have many more affective (dysthymia, major depression), dissociative (65 % with Child Dissociative Checklist scores over cutoff) and borderline personality features than males with similar severe trauma histories. The majority of all children in our experience who meet criteria for PTSD utilize a combination of persisting, adaptive responses present in the original trauma, involving an admixture of dissociative and hyperarousal responses. These diagnostic and phenomenological complications have hindered research and clinical practice with traumatized children (Scheeringa, Zeanah, Drell, and Larrieu, 1995; Terr, 1991).
There are a number of clear implications of a neurodevelopmental approach to the maltreated child (Appendix 1: Key Points: Clinical Work with Maltreated Infants). The first relates to the misunderstanding of resilience. We often hear "Children are resilient," or "They'll get over it, they didn't even know what was happening." It is not uncommon for adults to relate the traumatic events to clinicians in the presence of the child as if they were invisible. Often, recounting the event, the adults will describe how the traumatic event was terrifying for them, but as they describe the child's reactions they frequently misunderstand the child's unattached, nonreactive behaviors as 'not being effected' rather than the 'surrender' response. This pervasive, destructive view of caretaking adults in a young child's life exacerbates the potential negative impact of trauma. Of course, children 'get over it' -- they have no choice. Children are not resilient, children are malleable. In the process of getting over it, elements of their true emotional, behavioral, cognitive and social potential are diminished -- some percentage of capacity is lost, a piece of the child is lost forever.
Another major implication of a neurodevelopmental approach is that early intervention which can ameliorate the intensity and severity of the response to trauma will decrease the probability of developing, in a use-dependent fashion, sensitized neural systems resulting in either persisting hyperarousal or dissociative symptoms, or both. The more someone is in a dissociative state, the more likely they are to exhibit a dissociative symptomatology. The more they are in a fear state, the more likely they are to carry around persistent symptoms of hyperarousal.
The intensity and duration of response to trauma in children is dependent on a wide variety of factors. One of the most important appears to be the availability of a healthy and responsive caretaker to provide some support and nurturance for the child following the trauma. The presence of a healthy caretaker can diminish dramatically the alarm response or the dissociative response in the young child. Conversely, if the adult is impacted in a similar fashion by the trauma, that will have significant complicating impact on the child. Indeed, we have worked with a number of very young children and infants who, in acute traumatic experience (e.g., car accident), processed the experience very well. Primary caretaker who were with the child were significantly traumatized by the experience, however. Through their persisting anxiety and inability to contain their persisting hyperarousal symptoms, the primary caretakers, build into their children a mirroring hyperarousal fear response; a form of vicarious traumatization. And, if the 'caretaker' is the source of the trauma, the child's emotional survival depends on a zone of safety which includes new, true caretakers.
Different events at different times in the life of an individual are likely to result in a different combination of adaptive responses. An infant, a child and an adolescent experience the same event in different ways. What may be extremely traumatic for an adolescent may be a non- event for an infant (e.g., the threat of an armed robbery). Conversely, there are many experiences which are likely to be traumatic for infants which are minimally threatening for adolescents (e.g., separation from primary caretakers). Factors related to the individual's specific response to a given trauma include 1) pre-morbid functioning and history, -- specifically history of previous stressors, 2) age -- the neurobiological response patterns appear to change with age, 3) specific cognitive meaning of an event for an individual; 4) the specific nature of the trauma, and 5) presence of exacerbating (loss of caretaker) or attenuating factors (e.g., early intervention).
Children and infants use a variety of adaptive response patterns in the face of threat, and, in a use-dependent fashion, internalize aspects of these responses, organizing the developing brain. There are a variety of neuropsychiatric symptoms which result when these patterns of neural activation persist. This has implications for research, clinical assessment, intervention and prevention.
More important, however, is that understanding the impact of experience on the developing child by using a neurodevelopmental conceptualization offers certain directions for our culture (Perry, 1996). Profound socio-cultural and public policy implications arise from understanding the critical role of early experience in determining the functional capacity of the mature adult -- and therefore our society. Persistence of the destructive myth that 'children are resilient' will prevent millions of children, and our society, from meeting their true potential. Persistence of the pervasive maltreatment of children in the face of decreasing global and national resources will lead, inevitably, to sociocultural devolution.
It need not be so.
The clinical and clinical research work related to this paper has been supported, in part, by the CIVITAS Initiative and Mr. Alan Grant. Special acknowledgements are due to Andrew Vachss for helpful insights, commentary and editorial assistance on earlier versions of this manuscript.
Abercrombie, E. D., & Jacobs, B. L. (1988). Systemic naloxone administration potentiates locus coeruleus noradrenergic neuronal activity under stressful but not non-stressful conditions. Brain Research. 441, 362-366.
Abercrombie, E. D., Keefe, K. A., DiFrischia, D. S., Zigmond, M. J. (1989). Differential effects of stress on in vivo dopamine release in striatum, nucleus accumbens and medial frontal cortex. Journal of Neurochemistry 52, 1655-1658.
Altman, J., & Das, G. D. (1964). Autoradiographic examination of the effects of enriched environment on the rate of glial multipication in the adult rat brain. Nature. 204, 1161-1165.
Andrade, R., & Aghajanian, G. K. (1984). Locus coeruleus activity in vitro: Intrinsic regulation by a calcium-dependent potassium conductance but not alpha-2 adrenoceptors. Journal of Neuroscience. 4, 161-170.
Bhaskaran, D., & Freed, C. R. (1988). Changes in neurotransmitter turnover in locus coeruleus by changes in arterial blood pressure. Brain Research Bulletin. 21, 191-199.
Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the literature. Psychol Bulletin. 99, 66-77.
Cannon, W. B. (1914). The emergency function of the adrenal medulla in pain and the major emotions. American Journal of Physiology. 33, 356-372.
Coleman, P. D., & Riesen, A. H. (1968). Environmental effects on cortical dendritc fields: I. rearing in the dark. Journal of Anatomy (London!. 102, 363-374.
Cragg, B. G. (1975). The development of synapses in kitten visual cortex during visual deprivation. Experimental Neurology. 46, 445-451.
Cummins, R. A., & Livesey, P. (1979). Enrichment-isolation, cortex length, and the rank order effect. Brain Research, 178, 88-98.
DaCosta, J. M. (1871). On irritable heart: a clinical study of a form of functional cardiac disorder and its consequences. American Journal of Medical Science. 61, 17-52.
Davidson, J., & Smith, R. (1990). Traumatic experiences in psychiatric outpatients. J. Traum Stress. 3, 459-475.
DeBellis, M.D., Lefter, L. Trickett, P.K., Putnam, F.W. (1994) Urinary catecholamine excretion in sexually-abused giris. J Am Acad Child Adolesc Psychiatry. 33. 320-327.
DeBellis, M.D., Chrousos, G.P., Dorn, L.D., Burke, L. Helmers, K., Kling, M.A., Trickett, P.K. & Putnam, F.W. (1994) Hypothalamic-pituitary-adrenal axis dysregulation in sexually abused girsl. J. Clin. Endocrinol. Metab.. 78, 249-255.
Ebinger, P. (1974). A cytoachitectonic volumetric comparison of brains in wild and domestic sheep. Z Anat Entwicklungsgesch. 144, 267-302.
Elam, M., Svensson, T. H., & Thoren, P. (1984). Regulation of locus coeruleus neurons and splanchnic sympathetic nerves by cardiovascular afferents. Brain Research. 290, 281-287.
Famularo, R., Kinscherff, R., & Fenton, T. (1991). Posttraumatic stress disorder among children clinically diagnosed as borderline personality disorder. J Nerv Ment Dis. 179, 428431.
Giller, E. L., Perry, B. D., Southwick, S., & Mason, J. W. (1990). Psychoneuroendocrinology of post-traumatic stress disorder. In M. E. Wolf & A. D. Mosnaim (Eds.), Post-traumatic Stress Disorder: Etiology. Phenomenology and Treatment. (pp. 158- 170). Washington, D.C. American Psychiatric Press, Inc.
Glavin, G. B. (1985). Stress and brain noradrenaline: a review. Neuroscience Biobehavioral Review. 9, 233-243.
Goelet, P., & Kandel, E. R. (1986). Tracking the flow of learned information from membrane receptors to genome. Trends in Neuroscience. 9, 492-499.
Goldstein, D. S. (1995). Stress. Catecholamines and Cardiovascular Disease. New York: Oxford University Press.
Heinsbroek, R. P. W., van Haaren, F., Feenstra, M. P. G., Boon P., van de Poll N. E. (1991). Controllable and uncontrollable footshock and monoaminergic activity in the frontal cortex of male and female rats. Brain Research. 551, 247-255.
Henry, J. P., Stephens, P. M., & Ely, D. L. (1986). Psychosocial hypertension and the defence and defeat reactions. J Hypertension. 4, 687-697.
Henry, J. P., Liu, Y. Y., Nadra, W. E., Qian, C. G., Mormede, P., Lemaire, V., Ely, D., & Hendley, E. D. (1993). Psychosocial stress can induce chronic hypertension in normotensive strains of rats. Hypertension, 21, 714-723.
Herman, J. P., Guillonneau, R., Dantzer, R., Scatton, B., Semerdjian-Rouquier, L., Le Moal, M. (1982). Differential effects of inescapable footshocks and of stimuli previously paired with inescapable foootshocks on dopamine turnover in cortical and limbic areas of the rat. Life Science. 30, 2207-2214.
Hoffman, W. H., DiPiro, J. T., Tackett, R. L., Arrendale, R. F., & Hahn, D. A. (1989). Reiationship of plasma clonidine to growth hormone concentrations in children and adolescents. Journal of Clinical Pharmacology. 29, 538-542.
Ito, Y., Teicher, M. H., Glod, C. A., Harper, D., Magnus, E., Gelbard, H. A. (1993). Increased prevalence of electrophysiological abnormalities in children with psychological, physical, and sexual abuse. Journal of Neuropsychiatry and Clincal Neurosciences. 5, 401-408.
Kalivas, P. W. (1985). Sensitization to repeated enkephalin administration into the ventral tegmental area of the rat, ll: involvement of the mesolimbic dopamine system. Journal Pharmacolog Exp Ther. 235, 544-550.
Kalivas, P. W., Richardson-Carlson, R., & Van Orden, G. (1986). Cross sensitization between foot shock stress and enkephalin-induced motor activity. Biological Psychiatry. 21, 939-950.
Kalivas, P. W., Duffy, P., Dilts, R., Abhold, R. (1988). Enkephalin modulation of A10 dopamine neurons: a role in dopamine sensitization. Ann N Y Acad Sci. 537, 405- 414.
Kalivas, P. W., & Kuffy, P. (1989). Similar effects of daily cocaine and stress on mesocorticolimbic dopamine neurotransmission in the rat. Biological Psychiatry. 25, 913- 928.
Kandel, E. R., & Schwartz, J. H. (1982). Molecular biology of an elementary form of learning moduiation of transmitter release by cyclic AMP. Science. 218, 433443.
Kaufman, J. (1991). Depressive disorders in maltreated children. Journal of the American Academy of Child and Adolescent Psychiatry. 30 (2), 257-265.
Kiser, L., Heston, J., & Milisap, P. (1991). Physical and sexual abuse in childhood: relationship with post-traumatic stress disorder. Journal of the American Academy of Child and Adolescent Psychiatry. 30, 776-783.
Kleven, M. S., Perry, B. D., Woolverton, W. L., & Seiden, L. S. (1990). Effects of repeated injections of cocaine on D-1 and D-2 dopamine receptors in rat brain. Brain Research. 532, 265270.
Korf, J. (1976). Locus coeruleus, noradrenaline metabolism and stress. In E. Usdin, R. Kvetnansky, & I. J. Kopin (Eds.), Catecholamines and Stress. (pp. 105-111). New York: Pergamon.
Krystal, J. H., Kosten, T., Perry, B. D., Mason, J., Southwick, S., & Giller, E. L. (1989). Neurobiological aspects of post-traumatic stress disorder: review of clinical and preclinical studies. Behavior Therapy. 20, 177-198.
Lauder, J. M. (1988). Neurotransmitters as morphogens. Progress in Brain Research. 73, 365-388.
Leakey, R. (1994) The Origins of Humankind. Basic Books, New York.
LeDoux, J. E., Romanski, L., & Xagoraris, A., Romanski, L. M. (1989). Indelibility of subcortical emotional memories. Journal of Cognitive Neuroscience. 1, 238-243.
LeDoux, J. E., Cicchetti, P. O., Xagoraris, A., & et.al. (1990). The lateral amygdaloid nucleus: sensory interface of the amygdala in fear conditioning. Journal of Neuroscience. 10,1062- 1069.
Mason, J. W. (1971). A re-evaluation of the concept of 'non-specificity' in stress theory. Journal of Psychiatric Research. 8, 323-333.
McFarlane, A. C. (1987). Post-traumatic phenomena in a longitudinal study of children following a natural disaster. J Am Acad Child Adolesc Psychiatry. 26, 764-769.
Miczek, K. A., Thompson, M. L., & Tornatzky, W. (1990). Subordinate animals: Behavioral and physiological adaptations and opioid tolerance. In M. R. Brown, G. F. Koob, & C. Rivier (Eds.), Stress: Neurobioiogy and Neuroendocrinology. (pp. 323-357). New York: Marcel Dekker.
Moore, R. Y., & Bloom, F. E. (1979). Central catecholamine neuron systems: anatomy and physiology of the norepinephrine and epinephrine systems. Annual Reviews of Neuroscience. 2, 113-153.
Murberg, M. M., McFall, M. E., & Veith, R. C. (1990). Catecholamines, stress and posttraumatic stress disorder. In E. L. Giller (Ed.), Biological Assessment and Treatment of Post- traumatic Stress Disorder. (pp. 27-65). Washington, D.C. American Psychiatric Press, Inc.
Ogata, S. N., Silk, K. R., Goodrich, S., Lohr, N. E., Westen, D., Hill, E. M. (1990). Childhood sexual and physical abuse in adult patients with borderline personality diosrder. American Journal of Psychiatry.147,1008-1013.
Osofsky, J. D. (1995). The effects of exposure to violence on young children. American Psychologist (In press).
Perry, B. D., Stolk, J. M., Vantini, G., Gucchait, R. B., & U'Prichard, D. C. (1983). Strain differences in rat brain epinephrine synthesis and alpha-adrenergic receptor number: apparent 'in vivo' regulation of brain alpha-adrenergic receptors by epinephrine. Science. 221, 1297-1299.
Perry, B. D., Southwick, S. M., & Giiler, E. L. (1990a). Adrenergic receptors in post-traumatic stress disorder. In E. L. Giller (Ed.), Biological Assessment and Treatment of Post-traumatic Stress Disorder. (pp. 89-114). Washington, D.C. American Psychiatric Press.
Perry, B. D., Wainwright, M. S., Won, L., Hoffman, P., & Heller, A. (1990b). The influence of dopamine on dopamine receptor density in three dimensional tissue culture [Abstract]. Soc Neurosci Abstr.16. 646
Perry, B. D. (1994a). Neurobiological sequelae of childhood trauma: post-traumatic stress disorders in children. In M. Murberg (Ed.), Catecholamines in Post-traumatic Stress Disorder: Emerging Concepts. (pp. 253-276). Washington, D.C. American Psychiatric Press.
Perry, B. D., & Pate, J. E. (1994b). Neurodevelopment and the psychobiological roots of posttraumatic stress disorders. In L. F. Koziol & C. E. Stout (Eds.), The Neuropsychology of Mental Illness: A Practical Guide.
Perry, B.D. (1994c) Evolution of emotional, behavioral and physiological responses in children actutely exposed to violence. Presention at 41st Annual Meeting of the Academy of Child and Adolescent Psychiatry, New York.
Perry, B.D. (1995) Evolution of symptoms following traumatic events in children. [Abstract] Proceedings of the 148th Annual Meeting of the American Psychiatric Association, Miami.
Perry, B. D. (1996). Maltreated Children: Experience. Brain Development and the Next Generation. New York and London: W.W. Norton (in press).
Putnam, F.W. (1991) Dissociative disorders in children and adolescents: a developmental perspective. Psychiatric Clinics of NorthAmerica.14, 519-531.
Pynoos, R. S., Frederick, C., Nader, K., Arroyo, W., Steinberg, A., Eth, S., Nunez, F., & Fairbanks, L. (1987). Life threat and post-traumatic stress in school age children. Archives of General Psychiatry, 44, 1057-1063.
Sapolsky, R. M., Krey, L., & McEwen, B. S. (1984). Glucocorticoid-sensitive hippocampal neurons are involved in terminating the adrenocortical stress response. Proc Natl Acad Sci U S A. 81, 6174-6178.
Sapolsky, R. M., Uno, H., Rebert, C. S., Finch, C.E. (1990). Hippocampal damage associated with prolonged glucocorticoid exposure in primates. Journal of Neuroscience. 10, 2897-2902.
Scheeringa, M. S., Zeanah, C. H., Drell, M. J., & Larrieu, J. A. (1995). Two approaches to the diagnosis of post-traumatic stress disorder in infancy and early childhood. J Am Acad Child Adolescent Psychiatry. 34(2), 191-200.
Schwarz, E., & Kowalski, J. (1991). Malignant memories: post-traumatic stress disorder in children and adults following a school shooting. Journal of the American Academy of Child and Adolescent Psychiatry. 30, 937-944.
Schwarz, E. D., & Perry, B. D. (1994). The post-traumatic response in children and adolescents. Psychiatric Clinics of North America. in press
Selye, H. (1936). A syndrome produced by diverse nocuous agents. Nature. 138, 32-36.
Spik, R. A., & Wolf, K. M. (1946). Anaclitic depression: an inquiry into the genesis of psychiatric conditions in early childhood. ll. Psychoanalytic Study of the Child. (2), 313- 342.
Svensson, T. H. (1987). Peripheral, autonomic regulation of locus coeruleus noradrenergic neurons in brain: putative implications for psychiatry and psychopharmacology. Psychopharmacology. 92, 1-7.
Taylor, L., Zuckerman, B., Harik, V., & Groves, B. (1992). Exposure to violence among inner city parents and young children. American Journal of Diseases of Children. 146, 487
Teicher, M. H., Glod, C. A., Surrey, J., & Swett, C. ,Jr. (1993). Early childhood abuse and limbic system ratings in adult psychiatric outpatients. Journal of Neuropsychiatry. 5, 301- 306.
Terr, L. (1991). Childhood traumas: an outline and overview. American Journal of Psychiatry. 148, 1-20.
Terr, L. C. (1983). Chowchilla revisited: the effects of psychic trauma four years after a school-bus kidnapping. American Journal of Psychiatry. 138, 1543-1550.
Vantini, G., Perry, B. D., Gucchait, R. B., U'Prichard, D. C., & Stolk, J. M. (1984). Brain epinephrine systems: detailed comparison of adrenergic and noradrenergic metabolism, receptor number and 'in vivo' regulation in two inbred rat strains. Brain Research. 296, 49-65.