Published in the Journal of the American Academy of Child and Adolescent Psychiatry.
Reprinted with permission of the publisher.
Copyright (c)1992. All rights reserved.

Unlike other species such as rodents, strains of which have been bred for aggressiveness (Ebert, 1973; Lagersperz and Lagersperz, 1971), no particular ethnic, racial, or religious group has shown itself to be innately or enduringly more aggressive than any other (although from time to time throughout history, the people of one country or another have attempted to so distinguish themselves). The social and biological sciences have come to recognize that probably the most important influences on the development of violent behaviors are environmental or experiential. Although violence does not invariably beget violence (Widom, 1989), there is abundant evidence that a history of maltreatment is often associated with aggressive beahviors. From conception onwards, the ways in which living creatures are treated affect the ways in which they treat others of their species. Just how abusive treatment engenders aggression is not yet entirely understood. The purpose of this paper is to explore some of the most important psychobiological consequences of maltreatments in animals and humans and how they may contribute to aggression.

Aggression in animals and humand is not identical. Animal behavior tends to fall into two categories: affective and predatory (Flynn et al., 1970). Predatory aggression is accomplished with little autonomic activation: it involves carefully stalking and quickly subduing the prey in the interest of providing fast food. Affective aggression, on the other hand calls forth intense autonomic activation and includes among other behaviors intramale, fear-induced, and irritable aggression.

The nature of human aggression is not easily classified. The behaviors of Jack the Ripper and John D. Rockefeller have both been considered aggressive. For purposes of this discussion, human aggression or violence will be used simply to denote recurrent behaviors intended to cause pain, damage, or destruction to another person. It will not be used as a synonym for ambitiousness to assertiveness.

Environmental Influences on Aggression

Intrauterine Environment

The very position in utero of animals within a litter has been found to influence behavior. Female mice that develop in utero between two males are more aggressive after birth than those positioned between older females (vom Saal, 1984) long after delivery.

In humans, there is strong evidence that different kinds of noxious prenatal influences ranging from minor viral infections to maternal anxiety and psychological stress have been associated with childhood maladaptation (Pasamanick, 1956, 1961; Rutter, 1970; Stott, 1973; Stott and Latchford, 1976). The adverse effects of maternal alcoholism and other substances on fetal development and on subsequent postnatal social and intellectual functioning are well documented. The exposure of the fetus to abnormal levels of certain gonadal hormones has also been associated with behavioral sequelae (Ehrhardt and Money, 1967; Ehrhardt et al., 1968).

Stressful Living Conditions, Isolation, and Neglect

Overhearing, crowding, and uncomfortable living conditons have been demonstrated to promote aggression in animals (Griffit, 1970; Hutchinson, 1972). In humans, noxious odors (Jones and Bogat, 1978; Rotton et al., 1978), high temperatures, and increased population density (Griffit and Veitch, 1971) also have been shown to provoke hostility and aggression in experimental situations. These kinds of stressors are of special interest because of their possible implications for understanding some of the factors that contribute to aggression in impoverished, crowded urban environments.

Isolation of otherwise gentle animals during critical phases of development, an experience analogous in some ways to the neglect of human infants, is an especially powerful influence on the genesis of aggression behaviors (Brain and Nowell, 1971; Goldsmith et al., 1976; Luciano and Lore, 1976). The work of Spitz (1946), Bowlby (1969, 1975), and Provence and Lipton (1962) illustrates the devastating developmental consequences of isolation for human infants. Whereas almost total stimulus deprivation of infants can lead to extreme developmental delays, depression, and even death, lesser degrees of neglect have been associated with extremely poor peer relationships and with the development of aggressive behaviors (Cicchetti, 1989; Mueller and Silverman, 1989). In fact, as Widom (1989) has noted, there is a need to study neglect separately from abuse because of the evidence from certain studies that neglected children may even be more dysfunctional and aggressive than children who are physically abused. Noxious stimuli may actually be less detrimental to development than no stimulation at all.

Exposure to Aggressive Adults

The quality of early parenting also affects aggressiveness. Animals bred to have an especially gentle nature, if cross-fostered by adult females of a violent strain, will become more aggressive than is their usual nature (Southwick, 1968). There is evidence that mice reared in the company of their fathers as well as their mothers grow up to be more aggressive than those raised only by mothers (Mugford and Nowell, 1972). On the other hand, the results of cross-fostering aggressive strains of infant rodents with less aggressive adult animals are more variable. Certain strains of aggressive mice retain their aggressiveness even when reared by less aggressive adults (Southwick, 1968), whereas other strains do respond by becoming less aggressive (McCarty and Southwick, 1979; Smith and Simmel, 1977) when cross-fostered by less aggressive adults has a more powerful or predictable influence on subsequent adaptive styles than does exposure to more docile parent surrogates.

What are the psychological and behavioral consequences of aggressive parenting? We know that aggression can be learned. There is sound experimental evidence that modeling plays an important role in the development of aggressive behaviors both in animals (Hamburg, 1971) and in humans (Bandura, 1973). We also know that aggressive behaviors (like more adaptive behaviors) can be acquired through reinforcement. Patterson (1977) observed that when children's aggressive behaviors were ounished severely by parents, they tended to continue. On the other hand, when positive behaviors were reinforced by praise and aggressive behaviors given less attention, aggressive behaviors diminished (Patterson, 1979). Farrington (1978) also found severe physical punishment to be an antecedent of aggressive delinquency.

The author's own studies comapring extremely aggressive delinquents to their less aggressive delinquent peers indicated that the exposure to extremem violence within the household, particularly between caretakers, was strongly associated with children's violent behaviors (Lewis et al., 1979). When, however, in a follow-up of these delinquents (Lewis et al., 1989), an attempt was made to distinguish between those children raised in violent households and those who themselves were physically abused, it was found that the more information that was obtained, the clearer it became that abuse and exposure to other family violence tended to go hand in hand, making it difficult to assess the related importance of each.

Pain and Physical Abuse

Probably the most powerful genrator of aggression in animals and possibly in man is the repeated infliction of pain (Berkowitz, 1984). So strongly is this response in animals that a conventional experimental method for inducing murderous behaviors in mice and rats involves administering painful shocks to their feet (an ethically questionable practice). Physical torment also is an effective means of engendering viciousness in fighting dogs (e.g., pit bulls). The consequences of maltreatment in animals include the development of hypervigilance. Defeat in animals tends to engender defensiveness (Flannelly et al., 1984) which in turn is generalized to other situations and other opponents (Leschner, 1981; Seward, 1946).

Like animals, children who have been physically abused tend to behave in more aggressive ways than their nonabused peers (Cicchetti, 1989; Widom, 1989). These abused children have been noted to develop a hypervigilance, to misinterpret their surroundings, and, most importantly, to lash out when they perceive ambiguous stimuli as threatening (Dosge et al., 1984; Rieder and Cicchetti, in press). The author's studies of older children and adolescents revealed that the symptoms most characteristics of very violent youngsters were paranoid ideation and misperceptions, both of which are analogous to the hypervigilance of abuse children abd animals (Lewis et al., 1979, 1986). The findings of an association between paranoia and violence in delinquents was also consistent with the findings of Yesavage (1983a,b) who reported that the most important symptom distnguishing aggressive from nonaggresive psychiatric patients was paranoia.

Abusive treatment had been observed to have other deleterious effects that are peculiar to humans and that further contribute to aggression. Maltreatment affects expressive skills. Abused children have great difficulty putting their feelings into words (Cicchetti and Beeghly, 1987). Lacking the ability to convey emotions verbally, that tend, rather, to demonstrate their anger and misery through actions. Ironically, abused toddlers speak less about their negative feelings than do children from normal backgrounds. Cicchetti has hypothesized that such children develop overcontrolled styles of coping. However, there may be an alternative explanation for the clinical observations in this study suggest an alternative explanation. Severely abused children repress and totally deny abusive experiences and often do not even recall the abuse. The abused child's inability to identify, much less verbalize, his own distress probably accounts for his observed difficulty in appreciating the distress of others (Main and George, 1985). This lack of empathy noted in abused toddlers is most likely a reflection of their conditioned ability to insulate themselves from any stimuli that might reevoke their own painful experiences.

In summary, studies of human beings indicate that the same kinds of environmental stressors that increase violence in animals (i.e., intrauterine stressors, isolation or neglect, exosure to violent adults, and the infliction of pain) contribute to aggressive behaviors in humans. The question remains how much seemingly different kinds of experiences result in similar kinds of aggressive behavioral changes. Some possible psychophysiological mechanisms by which these kinds of experiences may engender aggression will be explored.

Physiological and Behavioral Consequences of Experiential Stressors

The fact that environmental or experiential influences affect the development of aggression does not mean that these effects are exclusively or even primarily psychodynamic or social. Children probably do not become recurrently violent simply as a result of modeling or reinforcement, although imitation and conditioned responses are contributory. Studies of animals and humans suggest that the environmental conditions and stressors that engender aggression are mediated at least in part physiologically.

Hormonal responses to Stressors

Environmental stressors affect hormone production. Levels of gonadal hormones are especially critical during fetal and neonatal growth (Goy and McEwen, 1980), influencing the development of sexually dimorphic areas of the brain (Hines, 1982). The presence of androgens prenatally has been shown to be crucial to the normal developments of aggressive behaviors in such diverse species as fish, lizards, birds, and chimpanzees (Floody and Pfaff, 1972). Androgens, which contribute to hypervigilance, are thought to sensitize the parts of the fetal brain that mediate aggression. Sensitized animals subsequently are able to respond rapidly and aggressively to stimuli that elicit surges of testosterone (Kamel et al., 1975). Studies have shown that when pregnant female rats are stressed by abnormal periodic exposure to bright lights, their serum testosterone levels as well as the testosterone levels of their male fetuses rise (vom Saal, 1984). Of note, episodic elevations of testosterone, especially if they occur at critical developmental stages for the fetus have been associated with increased postnatal aggressiveness. Thus, an environmental stressor affecting the mother has physiological consequences for the fetus and eventual behavioral consequences postnatally.

Some effects of prenatal exposure of humans to abnormally high levels of androgens have been observed in girls with congenital adrenal hyperplasia and in girls exposed in utero to exogenous masculinizing progestins. Their increased energy and athleticism, and their relative lack of interest in the more traditionally feminine concerns are thought to reflect an early in utero sensitization or masculinization of the brain (Ehrhardt et al., 1968; Ehrhardt and Money, 1967).

The early behavioral effects of gonadal hormones are not immutable, rather, hormones and their behavioral concomitants respond to the vicissitudes of life. In animal societies, such as monkey colonies, levels of circulating androgens respond to experiences of success or defeat and affect dominance and submissiveness. For example, when a previously dominant male is bested by another male, his testosterone may fall as much as 80%, and his place in the hierarchy of animals plummet (Rose et al., 1971). In contrast, successful challengers experience elevations of tesosterone and may assume more assertive roles in the colony.

One cannot simply extrapolate from the effects of gonadal hormones on animals to their effects on man. There is some evidence that experiences of success, as in winning a tennis match or graduating from medical school, are associated with elevation of serum testosterone levels in men (Mazur and Lamb, 1980). However, for the most part, results of studies regarding the relationship between testosterone levels and aggression in humans have been equivocal (Ehrenkraz et al., 1974; Kreuz and Rose, 1972; Meyer-Bahlburg, 1974; Monti et al., 1977; Rada et al., 1976). The use of anabolic steroids in body builders has been associated with extremes of aggressive behavior (Pope and Katz, 1988) and with the development of paranoid symptoms (Pope and Katz, 1987; Tennant et al., 1988). Of note, although diminishing levels of circulating androgens in men by means of chemical or surgical castration has been found to reduce sexual aggression (i.e., rape) (Tupin, 1987), it has not been demonstrated to be an effective biological treatment for the suppression of other kinds of nonsexual violent behavior. Its effectiveness as a deterrent, no doubt, would be considerable.

Gonadal hormones are not the only hormones affected by environmental stressors. Defeat, which tends to lower testosterone levels causes circulating corticosteroids to rise (Christian, 1955, 1959). When an animal has been repeatedly exposed to defeat, even exposure to a potential aggressor will cause a surge of corticosteroids (Bronson and Elftheriou, 1965a, b). Similarly, the adminstration of high doses of corticosteroids to a dominant male will induce submissive behavior (Flannelly et al., 1984).

The relationship of endocrine status to behavior is complex and far from fully understood. The effects of angdrogens, estrogens, progestins, and corticosteroids, as well as myriad other hormones, vary tremendously. The discussion of testosterone and corticosteriods is but an illustration of some of the ways in which experiences can affect behavior. These interactions demonstrate the important fact that the biological state of the organism, animals or man, is not fixed but is, rather, continually changing and responding to the environment. Thus what may appear to be an inherent aggressive temperament can, in fact, be the reflection of a physiological state induced and reinforced by environmental stressors.

Neurotransmitter Responses to Stressors

Just as environmental stressors affect hormones levels, so there is an ever growing literature documenting the effects of stress on brain levels of neurotransmitters. What is more, studies of animals and humans suggest that certain neurotransmitters such as norepinephrine, dopamine, and serotonin play important roles in both the genesis and suppression of aggressive behaviors.

For example, aggressive beahviors in animals have been induced by administering precursors of norepinephrine and by administering noradrenergic mimetic drugs (Lal et al., 1968, 1970; Randrup and Munkvad, 1966; Reis, 1972). Although the ways in which norepinephrine enhances aggression are not known, it is thought that its presence in specific parts of the brain inhibits certain brain stem neurons known to suppress aggression (Reis, 1972). Thus, it is thought to inhibit inhibitors. (It is important to recall that there are different kinds of aggression in animals. Norepinephrine has been found to facilitate affective or intermale aggression while inhibiting the predatory behaviors associated with food acquisition. Similarly, substances that increase intermale aggression suppress infanticidal behaviors [vom Saal, 1979]. Thus, when considering behavioral effects on animals of particular neurotrnasmitters, it is essential that the type of aggression be specified.)

Dopamine is also thought to play a role in the affective or intermale aggression of animals. For example, when mice are given methylparatyrosine, an inhibitor of dopamine synthesis, their aggressiveness diminishes (Lycke et al., 1969). The behavioral effects on humans of fluctuations in norepinophrine and dopamine are uncertain. Whereas the calming, focusing effects on stimulant medication suggest that increases in central nervous system dopamine and norepinephrine diminsh hyperactivity and consequently lessen aggressive behaviors, the tranquilizing effects of the antipsychotic medications that block dopamine receptors suggest an opposite behavioral effect of dopamine. In short, at this time, it is impossible to generalize about the role of these kinds of neurotransmitters on human aggression.

One of the most studied neurotransmitters in terms of its influence on aggressive behaviors in animals and man is serotonin. For example, Sahakian (1981) reported low concentrations of serotonin in the cerebrospinal fluid of hyperaggressive rats. Others have demonstrated inceased aggression in mice and rats whose brains for one reasons or another have been depleted of serotonin (Alpert et al., 1981; Lycke et al., 1969). When mice are isolated at critical developmental stages, their concentration of brain serotonin diminishes, and these serotonin-depleted mice become aggressive (Garattini et al., 1969; Valzelli, 1974). Conversely, substances that potentiate serotonin have been demonstrated to diminish aggression. (Alpert et al., 1981).

In humans there is a growing body of literature suggesting a relationship between diminished levels of brain serotonin and self-injurious (Asberg et al., 1976, 1987) as well as outwardly aggressive behaviors (Brown et al., 1979). Based on studies of violent offenders and impulsive fire setters, Linnoila and colleagues (1983) suggested that low cerebral scrotonin and concentration (as reflected in low CSF-5-HIAA) may be associated with impulsive beahviors in general rather than aggressiveness or violence in particular. In a follow-up study of these subjects, Virkkunen and his colleagues (1989) reported diminished concetrations of both 5-HIAA and HVA (metabolites of serotonin and dopamine) in recidivists as comapred to nonrecidivists. Recently, Coccaro and colleagues (1989) demonstrated diminished prolactin responses to fenfluramine (a serotonin-releasing agent) in impulsive-aggressive patients diagnosed as having personality disorders and in depressed suicidal patients. Consistent with these data is fact that many of the antidepressant medications increase serotonin and relieve agitation and suicidal behavior. Coccaro and colleagues (1989) hypothesized the existence of a psychobiological suspecibility to impulsive-aggressive behavior, secondary to insufficient serotonin in the brain. This theory, however, does not explain why some individuals with diminished amounts of serotonin simply become unhappy and irritable whereas others become aggressive. What else is necessary to cause a person in a serotonin-depleted, irritable state to aggress against another person rather than bite his own nails or tear at his own flesh?

Brain Function and the Interaction of Hormones and Neurotransmitters

Neuroanatomy, hormones, and neurotransmitters cannot really be studied separately. A few examples of their interactions are offered:

1. In vitro studies of the effects of hormones on the embryonic rat brain show that both estrogen and progesterone enhance the growth of mesencephalic dopamine neurons (Reisert et al., 1987). Thus, hormones actually affect brain structure.

2. Studies of rats and hamsters reveal that hormones associated with aggression, such as estradiol and testosterone, concentrate in specific limbic system nuclei (Floody and Pfaff, 1972).

3. The introduction of testosterone into particular hypothalamic nuclei of castrated rats causes them to resume those aggressive beahviors that were previously suppressed by castration (Herbert, 1989). In this study testosterone seems to act as a neurotransmitter.

4. In vitro studies of the effect of neurotransmitters on testicular tissue in hamsters reveal the role of catecholamine in the production of testosterone (Maycrhofer et al., 1989).

5. Serotonin agonists, releasers, precursors, and uptake inhibitors have been found to elevate corticortisone levels in rats (Fuller, 1981).

Clearly the actions of hormones cannot be studied without considering their interactions with neurotransmitters and vice versa.

Where in the brain do the neurophysiological interactions related to aggression occur? Clinical data and data from experimentation with animals have shown that three areas of the brain are especially important in terms of the modulation of violent behaviors. MacLean (1985) and Weiger and Bear (1988) have conceptualized a hierarchy of neural controls that involves particularly the hypothalamus amygdala and orbital prefontral cortex. The hypothalamus, which receives input from osmo and chemoreceptors and from the amygdala, affects endocrine responses through its influence on the pituitary. For example, damage to the anterior hypothalamus of male rats reduces sexual and aggressive behaviors (Floody and Pfaff, 1972).

In addition, hypothalamus projections to the brain stem control stereotyped behaviors, a phenomenon illustrated by the sham rage of decorticate cats (Bard, 1928). Aggressive behaviors can be elicited or suppressed, depending on which parts of the hypothalamus are stimulated or abated. Aceytlcholine has been shown to be an important neurotransmitter in this area of the brain (Bandler, 1970; Bear et al., 1986; Smith et al., 1970). In humans, leasions in the hypothalamus have been associated with unplanned animal-like attacking behaviors.

In contrast with the hypothalamus, the amygdala receives input from all sensory modalities. It has projections to the hypothalamus and plays a role in the association of particular sensory stimuli with aggressive responses (Downer, 1961; Weiger and Bear, 1988). Lesions in the amygdala have been shown to impair an animal's ability to distinguish between appropriate and inappropriate objects for satisfying hunger and sexual drives (Keating, 1971; Kluver and Bucy, 1939).

Experiments on animals suggest that stimulation of the amygdala, located deep within the temporal lobe, is involved in the kind of aggression that occurs in response to fear (Egger and Flynn, 1963; Siegel and Flynn, 1968). On the other hand, damage to the amygdala has been reported to result in a diminution of aggressive behaviors in response to novel stimuli. In humans, lesions in the amygdaloid area have been associated with apathy and hyposexuality, whereas abnormal electrical activity in this area has been associated with aggression. Whether or not directed aggression ever occurs during and actual seizure originating in the amygdala or in other limbic structures of the brain remains a topic of debate.However, interictally, many patients with epileptic disorders experience irritability, intensification of feelings, fearfulnees, and outright paranoia (Bear and Fulop, 1987; Lewis, 1976; Lewis et al., 1982).

This finding--that a particular part of the brain involved in aggression is especially responsive to stimuli that elicit fear--is important because, as previously discussed, paranoid misperceptions, i.e., fearfulness, hypervigilance, and an unwarranted sense of threat, play such an important role in the etiology of violent behavior in human beings. It is during these kinds of emotionally intense interictal periods that planned, purposeful interpersonal violence can occur. It is important to recognize that intentionally violent behaviors can and do occur as a result of brain dysfunction and that premeditation is not necessarily an indication of sanity.

The frontal cortex, that part of the brain so essential for abstract thought, planning, and judgement, interacts with the rest of the neocortex as well as the amygdala and hypothalamus. Lesions of the frontal cortx have been associated with apathy, impulsivity, and irritability, depending on their location. Damage to the dorsolateral convexity has been linked with apathetic, irresponsible behaviors, whereas damage to the orbital area is more commonly linked with impulsive, inappropriate beahviors (Blumer and Benson, 1975; Luria, 1980).

Although attempts have been made to explain specific kinds of criminally aggressive behaviors in terms of the localizations of lesions, as yet there are no hard and fast rules that apply invariably to specific human behaviors. The above examples of brain function and dysfunction are, in fact, gross oversimplifications. No part of the brain works in isolation, and, as noted, the structures involved in aggression have widespread connections to other parts of the brain. The essential principle to keep in mind is that in real life the expression of violence is not simply the outcome of localized stimulation. Different parts of the brain are continuously interacting with each other, and violent behaviors reflect the result of the equilibrium achieved between the stimulation and suppression of particular areas of specific points in time.

Are All People Created Equal in Terms of Aggressive Responses to Maltreatment?

Are we all created equal or are some of us innately more aggressive than others? Studies of infants indicate that we do not enter the world as temperamentally identical blank slates. Some infants are more tense and irritable than others and are more given to temper tantrums when stressed by the ordinary vicissitudes of life (Chess and Thomas, 1984). The same squeaky chalk will feel more abrasive to some slates than others. However, early temper tantrums are not predictive of a life of violent crime (Kagan and Moss, 1962) and, although aggressive behaviors subsequent to age 3 years are often associated with ongoing interpersonal problems, the majority of aggressive young children do not become violent adults.

Are there any genetic abnormalities that impart a specific tendency toward vioelnce? With the exception of Losch Nyhan Syndrome (Palmour, 1983), no genetic abnormality has yet been identified that predisposes an individual specifically to violent behavior.

The findings reported during the 1960s and 1970s suggesting that certain chromosomal abnormalities (e.g., XYY and XXY anomalies) were associated with a predisposition to violence. (Casey et al., 1966; Forssman and Hambert, 1967; Hook, 1973; Nielson, 1968; Telfer, 1968; Witkin et al., 1976) have been reevaluated, and their conclusions questioned (baker et al., 1970; Gerald, 1976; Jacobs et al., 1971; Schiavi et al., 1984). Most probably, these abnormal constellations of chromosomes, like other abnormal conditions, predispose an individual to a variety of different kinds of adaptational problems which, depending on upbringing and stressors, may or may not be manifested by aggression.

The only chromosomal constellation or syndrome that has repeatedly been demonstrated to be associated with aggressive behavior is the "XY Syndrome." A major feature of the XY Syndrome is a diminished violence threshold, and this characteristic is true of most animals as well as man. What is it about the male consition that creates this tendency to respond aggressively? Given the fact that this quality of temperament is not peculiar to humans, but rather is equally characteristic of animals, it makes sense to conclude that physiological rather than simply societal influences are at play. It would seem that boys with their testosterone-sensitive masculinized brains and their physiological capacity to secret large amounts of androgen in response to particular stimuli, are, from the outset, more susceptible than girls to the aggression-promoting effects of maltreatment.

What other physiological conditions or states lower the threshold for violent responses to stressful stimuli? After all, not all abused children, not even all abused boys, become violent. What makes one boy more susceptible than the next to the "slings and arrows of outrageous fortune?"

The author's studies of violent individuals suggest that some children are indeed more susceptible than others to the violence engendering effects of abuse and neglect. For example, the studies found that among abused boys, those with psychiatric, neurological, and cognitive impairments are far more likely to act aggressively than those whose central nervous system functions are intact (Lewis et al., 1989). Severely neurophysiologically impaired girls, on the other hand, even when abused are far less violent than boys (Lewis et al., 1991). It seems reasonable, therefore, to conclude that impaired boys, who for any number of reasons already have neurophysiological vulnerabilities, who either start with abnormal neurotransmitter concentrations or abnormal physiological responses to stress, are more likely than their impaired female counterparts or their more neuropsychiatrically intact male counterparts to be affected adversely by the psychobiological consequences of abuse or neglect. The boy who is already intrinsically vulnerable to paranoid misperception by virtue of an inherent neurophysiological predisposition to major psychiatric illness, or the boy who suffers from some sort of central nervous system dysfunction that increases irritability and impulsivity, might be expected to respond especially aggressively to the additional psychological and neurophysiological insults imposed by neglectful or abusive treatment.

We do not know that once a physiological response has been established it is easily revoked by exposure to similar stimuli. It is, therefore, reasonable to hypothesize that ongoing abuse and neglect, especially in early childhood, have conditioning effects, setting up the kinds of neurophysiological circuits and results in recurrent aggressive behavior.

What are the most common characteristics of violent individuals? The literature has tended to focus separately on different concomitants of aggression. Some have concentrated primarily on impulsivity and the biochemical abnormalities associated with it; others have stressed the importance of paranoia in repeatedly antisocial individuals; still others have focused on cognitive or intellectual deficits. In spite of these findings, it is important to remember that most brain-damaged, impulsive children are not violent; and most cognitively impaired, learning disabled, or retarded children are not violent. These separate vulnerabilities do not, in and of themselves, seem to create violence. Furthermore, most abused children do not turn into violent criminals (Widom, 1989).

On the other hand, work by the author over the years has suggested that when neuropsychiatric and cognitive deficits exist together, maltreatment is an especially potent precipitant of aggression. That is, when impulsivity, hypervigilance, and cognitive expressive deficits coexist, the psychophysiological stage is set for violence to occur. What is it about these kinds of vulnerablities that create a matrix for violence? First, brain dysfunction of almost any kind is often associated with irritability, impatience, and moodability. Second, paranoid ideation and misperceptions, symptoms associated with so many different kinds of psychiatric disorders, increase fearfulness and a tendency to retaliate for both genuine and imagined threats. Finally, cognitive deficits not only impair judgement but also diminish the ability to conceptualize feelings and put them into words rather than actions.

How might one understand the ways in which maltreatment in the form of abuse or neglect exacerbates these vulnerabilities and encourages violence? Certainly, maltreatment has psychodynamic consequences, engendering rage and providing a model of violent behavior. In addition, extrapolating from the research on animals, maltreatment modifies the physiology of the organism itself. It is reasonable to hypothesize that abusive, neglectful treatment diminishes concentrations in the brain of substances such as serotonin that ordinarily help to modulate feelings, maltreatment seems to increase the outpouring of substances such as dopamine and testosterone that enhance competitive and retaliatory aggression. These same substances also contribute to hypervigilance, and thus increase the fearfulness and paranoia that give rise to violent acts.

There is also another way in which these kinds of physiological reactions to maltreatment may possibly contribute to aggression. We know that testosterone and other hormones affect the very structure of the brain. The delayed verbal skills of boys compared to girls is thought to be a reflection in part of the action of testosterone on the developing brain (Hines, 1982). The special difficulties that abused toddlers have expressing feelings in words may not be simply a reflection of psychological intimidation but rather a manifestation of neuroanatomical and neurophysiological changes secondary to abusive or neglectful treatment. Furthermore, the apparent lack of empathy of abused aggressive children noted previously may be a manifestation of centrally medicated expressive deficit coupled with a conditioned imperviousness to certain painful stimuli and not simply a reflection of nastiness or character pathology. Studies of abused young children indicate that they do not suffer from a lack of moral development. Abused children, as well as normal children, consider it wrong to cause physical harm (Smetana and Kelly, 1989). It would seem, rather, that abused children are unable to act on their intellectual understanding of moral principles when they are stresses. When such traumatized children also happen to be paranoid (which is frequently the case), they tend to adopt defensive, contemptuous attitudes toward the rest of the world, thus seeming to complete the picture of sociopathy.

In short, whatever increases impulsivity and the irritability engenders hypervigilance and paranoia, diminishes judgement and verbal competence, and curtails the ability to recognize one's own pain and the pain of others, also enhances the tendency toward violence. Abusive, neglectful caretaking does all of these things. In a resilient child, maltreatment (i.e. abuse or neglect) may not engender aggression. In an already vulnerable child with tendencies toward impulsivity, hypervigilance, expressive difficulties, and dissociation from painful feelings, maltreatment is often sufficient to create a very violent individual. To the extent that testosterone contributes to this constellation of vulnerabilities, persons with the XY Syndrome are at special risk.

What are the implications of the above psychobiological phenomena for the treatment and prevention of violence? Clearly, interventions that help control impulsivity, diminish irritability, enhance a child's sense of security, alleviate paranoid feelings, improve cognition and verbal expressiveness, and encourage recognition of one's own pain and the pain of others,will diminish the likelihood of violence. To date, programs addressing each of these issues individually and specifically do not exist.

On the other hand, ironically, our correctional system produces all of the ingredients known to promote violence: isolation, discomfort, exposure to other aggressive individuals, insecurity, and lack of intellectual stimulation. If in our prisons we have demonstrated our ability to make a laboratory which predictably produces and reinforces aggressive behavior, surely the possibility exists that with a little ingenuity we might, just as reliably, be able to create an environment to produce and encourage the opposite.

---

References:

Alperi, J.E., Cohen, D.J. Shaywitz. B,A, & Piccicilo, M. (1981), Neurochemical and beahvioral organization: disorders of attention, activity and aggression. In: Vulnerabilities to Delinquency, ed. D.O. Lewis. New York: Spectrum, pp.109-171.

Asberg, A., Traskman. L. & Thoren. P. (1976), 5-HLAA in cerebrospinal fluid: a biochemical suicide predictor? Arch. Gen. Psychiatry 33:1193-1196.

Asberg, M., Scalling, D., Trakeman-Bendz I., & Wagner, A. (1987), A psychology of suicud, impulsivity, and related phenomena. In: Psychopharmacology: Third Generation of Progress, ed. H.Y. Meltzer. New York: Raven Press. pp. 655-688.

Baker, D., Telfer M.A., Richardson, C.E. & Clark, G.R. (1970). Chromosome errors in men with antisocial behavior: comparison of selected men with "Klinefelter's syndrome" and XYY chromosome pattern. JAMA, 214:869-878.

Bandler, R.J. (1970), Cholinergic synapses in the lateral hypothalamus for the control of predatory aggression in the rat. Brain Res. 20:409-424.

Bandura, A. (1973). Agggression: A Social Learning Analysis. Englewood Cliffs, NJ: Prentice Hall.

Bard, P. (1928). A diencephalic mechanism for the expression of rage with special reference to the sympathetic nervous system. Am. J. Physiol. 84:490-515.

Bear, D. M. & Fulop, M. (1987). The neurology of emotion. In: Behavioral Biology in Medicine, ed. A. Hobson, South Norwalk CT: Meduration, Inc.

--Rosenbaum, J.F. & Norman, R. (1986). Aggression in cat and man precipitated by a cholinesterade inhibitor, Psychosomatics 26:535-536.

Berkowitz, L. (1984) Physical pain and the inclination to aggression. In: Biological Perspectives on Aggression, eds. K.J. Flannelly, R.J. Blanchard & D.C. Blanchard. New York: Liss. pp.27-47.

Blumer, D. & Benson, D.F. (1975), Personality changes with frontal and temporal lobe lesions. In: Psychiatric Aspects of Neurologic Disease, eds. D.F. Benson & D. blumer, New York: Grune and Stratton. pp.151-170.

Bowlby, J. (1969). Attachment and Loss. vol.1. New York: Basic Books.

--(1975). Attachment and Loss, vol.2. Harmondsworth: Penguin. Brain, P.F. & Nowell, N.W. (1971). Isolation versus grouping efffects on adrenal and gonadal function in albino mice. 1. The male. Gen. Comp. Endocrinol., 16:1489.

Bronson, P.H. & Eleftheriou, B.E. (1965a), Adrenal response to fighting mice: separation of physical and psychological causes. Science 147:627.

--(1965b), Relative effects of fighting on bound and unbound corticosterone. Proc. Soc. Exp. Biol. Med. 118:146.

Brown, G.L., Ballenger, J.C., Minichiello, M.D. & Goodwin, F.K. (1979). Human aggression and its relationship to cerebrospinal fluid 5-hydroxy-indoleacetic acid, 3-methoxy-4-hydroxy-phenyl-glycol, and homovanillic acid. In: Psychopharmacology of Aggression, ed. M. Sandler. New York: Raven Press. pp. 131-148.

Casey, L.J., Seagall, D.R., Street, K. & Blank, C.E. (1966). Sex chromosomes abnormalities in two state hospitals for patients requiring special security. Nature, 209:641-642.

Chess, S.E.& Thomas, A. (1984), Origins and Evolution of Behavior Disorders, New York: Brunner and Mazel.

Christian, J.J. (1955). Effects of population size on the adrenal glands and reproductive organs of male mice in populations of fixed size. J. Physiol. (Lond.) 182:292.

--(1959), Lack of correlation between adrenal weight and injury from fighting in grouped male albino mice. Proc. Soc. Exp. Biol. Med. 101:166.

Ciccheiti, D. (1989), How research on child mattreatment has informed the study of child development; perspectives from developmental psychopathology. In: Child Maltreatment: Theory and Neglect, cds. D Chicchetti & V. Carlson. New York: Cambridge University Press, pp. 377-431.

--Beeghly, M. (1987), Symbolic development in maltreatment youngsters: an organizational perspective. In: Atypical Symbolic Developmental. eds. D. Cicchetti & M. Beeghly. San Francisco, Iossey-Bass.

Coccaro, E. F., Siever. L. J., Klar, H. M., Maurer, G. et al. (1989). Serotonergic studies in patients with affective and personality disorders. Arch. Gen. Psychiarry. 46:587-598.

Dodge. K, Murphy, R.& Buchsbaum, K. C. (1984). The assesment of intention-cue detection skills in children:implications for developmental psuchopathology. Child Dev.55:163-173.

Downer, J. L. (1961). Changes in visual gnostic functions and enotional behaviour following unilateral temporal pole damage in the "split-brain" monkey, Nature, 191:50-51.

Eberr, P. D. (1983). Sclection for aggression in a natural population. IN; Aggresive Behavior:Genetica and Neural Approaches. eds. E. C. Himmel. M.E. Hahn & J.K. Walters. Hillsdale. NJ:Laurence Erlbaum, pp. 103-27.

Egger. M.D.& Flynn. J. P. (1963). Effects of clectriacal stimulation of the amygdala on hypothalmically elicited attack behavior in cats. J, Neurophysiol., 26:705-720.

Ehrenkranz. J., Bliss, E. & Sheard, M. H. (1974), Plasma testosterone: correlation with aggresive behavior and social dominace in ma. Psychoson. Med. 36:469-475.

Ehrhardt, A. A. & Money, J. (1967). Progestin-induced hermaphrodictism: IQ and psychosezual identity in a samle of ten girls. Journal of Sex Research, 3:83-100.

--Epstein, R. & Money. J. (1967), Fetal androgens and female gender identity in the early treated adrenogenital syndrome. John Hopkins Medical Journal. 122:160-167.

Farrington, D. P. (1978), The family backgrounds of aggressive youths. In: Agression and Antisocial Behavior in Childhood and Adolescence, eds. L. A. Hesov, M. Berger & D. Shaffer. Oxford. Pergarnon. pp. 73-93.

Flannely, K. J., Flannelly, L. & Blanchard, R. J. (1984). Adult eperience and the expression of aggression: a comparative analysis. In: Biological Perspectives on Aggression, eds. K. K. Flannely, R, J. Blanchard & D.C. Blanchard. New York: Liss, pp.207-259.

Floody, O. R. & Pfaff, D, W, (1972). Steriod hormones and aggressive behavior: approaches to the study of hormone-sensitive brain mechanisms for behavior. S. H. Frazier. Aggression, 52:149-184.

Flynn. J. P., Vanegas. H.. Foote, W. & Edward, S. (1970). Nueral mechanism involved in a cat's attack on a rat. In: Nueral Control of Behavior, eds. R. E. Whalen et al., New York: Academic Press. p. 135.

Forssman, H. & Hamber, G. (1967). Chromosomes and antisocial behavior. Exerpta Criminologica, 7:113-117.

Fuller, R. W. (1981) Segrotonergic stimulation of pituitary-adrenocortieal functions in rats. Nwuroendocrinology. 32:118-127.

Garattini, S., Giacolone, E. & Balzelli, L. (1969). Biochemical changes during isolation-induced aggressiveness in mice. In: Agressive Behavior, eds. S. Garattini & E. Sigg. New York: Wiley.

Gerald, P. S. (1976). Current concepts in genetics: sex chromosome disorders. New England Journal of Medicine, 294:706.

Goldsmith, J. F., Brain. P. F. & Benton. D. (1976). Effects of age at differential housing and the duration of individual housing/grouping on intermale fighting behavior and adrenocorrical activity in T. O. strain mice, Aggressive Behavior, 2:307-323.

Goy, R. W. & McEwen, B. S. (1980). Sexual Differentiation of the Brain. Cambridge, MA: MIT Press.

Griffitt, W. (1970). Environmental effects on interpersonal affective behavior; ambient effective temperature and attraction. J. Pers. Soc. Psychol. 15:240.

--Veitch, R. (1971). Hot and crowded, influence of population density and temperature on interpersonal affective behavior: ambient effective temperature and attraction. J. Pers. Soc. Psychol. 17:92-98.

Hamburg, D.A. (1971), Psychological studies of aggressive behaviour. nature. 230:19-23.

Herbert, J. (1989), The physiology of aggression, In: Aggression and War: Their Biological and Social Bases. eds. J. Groebel & R.A. Hinde. New York: Cambridge University Press, pp.58-71.

Hine, M. (1982). Prenatal gonadal hormones and sex differences in human behavior. Psychol. Bull. 92(1):56-80.

Hook, E.B. (1973). Behavioral Implications of the human XYY genotype. Science, 179:139-150.

Hutchinson, R.R. (1972). The environmental causes of aggression. In: Nebraska Symposium on Motivation, eds. J.K. Cole & D.D. Jensen. Lincoln: University of Nebraska Press.

Jacobs, P.A., Price, W.H., Richmond, S. & Ratcliff, B.A.W. (1971). Chromosomes surveys in penal institutions and approved schools. J. Med. Genet. 8:49-58.

Jones, J.W. & Bogat, G.A. (1978). Air pollution and human aggression. Psychol. Rep. 43:721.

Kagan, I. & Moss, H. (1962). From Birth to Maturity. New York:John Wiley.

Kamel, F., Mock, E.J., Wright, W. W. & Frankel, A.I. (1975). Alterations of plasma concentrations of testosterone, LH, and prolactin associated with mating in the male rat. Horm. Behav. 6:277.

Keating, E.G. (1971). Somatosensory deficit produced by parietotemporal disconnection. Anat. Rec. 169:353-354.

Klover, H. & Bucy, P.C. (1939). Preliminary analysis of functions of the temporal lobes in monkeys. Archives of Neurology and Psychiatry 42:979-1000.

Kreuz, L.E. & Rose, R.M. (1972). Assessment of aggressive behavior and plasma testosterone in a young criminal population. Psychosom. Med. 34:321-332.

Lagerspetz, K.M.J. & Lagerspetz, K.Y.H. (1971). Changes in the aggressiveness of mice resulting from selective breeding, learning and social isolation. Scand. J. Psychol. 12:241-248.

Lal, H. De Foo, J.J. & Thut, P. (1968). Effect of amphetamine on pain-induced aggression. Communicaations in Behavioral Biology. 1:333.

--Nesson, B. & Smith, N. (1970). Amphetamine-induced aggression in mice pretreated with dihydroxyphenylalanine (DOPA) and/or reserpine. Biol. Psychiatry. 2:299.

Leshner, A.L. (1981). The role of hormones in the control of submissiveness. In: A Multidisciplinary Approach to Aggression Research. eds. P.F. Brain & D. Benton. Amsterdam: Elsevier/North Holland Press. pp. 309-322.

Lewis, D.O. (1976). Delinquency, psychomotor epileptic symptoms and paranoid ideation: a triad. Am. J. Psychiatry. 133:1395-1398.

--Shanouk, S. Pincus, J. & Glaser, G. (1979). Violent juvenile delinquents: psychiatric, neurological, psychological and abuse factors. J. Am. Acad. Child Adolesc. Psychiatry. 18:307-319.

--Pincus, J., Shanok, S. & Glaser, G. (1982). Psychomotor epilepsy and violence in a group of incarcerated adolescent boys. Am. J. Psychiatry. 139:882-887.

Lovely, R., Yeager, C. & Della Femina, D. (1989). Toward a theory of the genesis of violence: a follow-up study of delinquents. J. Am. Acad. Child Adolesc. Psychiatry 28:431-436.

--Pincus, J.H., Feldman, M., Jackson, L. & Bard, B. (1986). Psychiatric, neurological, and psychoeducational characteristics of 15 death row inmated in the United States. Am. J. Psychiatry. 143:838-845.

LOvely, R., Yeager, C. & Della Fermina, D. (1989). Toward a theory of the genesis of violence: a follow-up study of delinquents. J.Am. Acad. Child Adolesc. Psychiatry 28:431-436.

--Pincus, J.H., Feldman, M., Jackson, L. & Bard B. (1986). Psychiatric, neurological, and psychoeducational characteristics of 15 death row inmates in th Unted States. Am J. Psychiatry, 143:838-845.

--Yeager, C., Cobham-Portorreal, C.S., Klein, N., Showalter, C. & Anthony, A. (1991). A follow-up female delinquents, maternal contributions to the perpetuation of deviance. J. AM. Acad. Child Adolesc. Psychiatry. 30:197-201.

Linnoila, M., Virkkunen, M., Scheinin, M., Nuutila, A., Rimon, R., & Goodwin, F.K. (1983), low cerebrospinal fluid 5-hydroxyindoleascetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Sci. 33:2609-2614.

Luciano, D. & Lore, R. (1975). Aggression and social experience in domesticated rats. J. Comp. Physiol. Psychol. 88:917.

Luria, A.R. (1980), Higher Cortical Functions in Man. New York:Basic Books.

Lycke, E., Modigh, K. & Roos, B.E. (1969). Aggression in mice associated with changes in the monoamine-metabolism of the brain, Experientia 25:951-953.

MacLean, P. (1985). Brain evolution relating to family, play and the separation call. Arch. Gen. Psychiatry 42:405-417.

Main, M. & George, C. (1985), Response of abused and disadvantaged toddlers to distress in agitates: a study in the day care setting. Developmental Psychology, 21:407-412.

Nayerhofer, A. bartke, A. & Steger, R.W. (1989), Catecholamine effects on testicular testosterone production in the gonadally active and the gonadally regressed adult golden hamster. Biol. Reprod. 40:752-761.

Mazur, A. & Lamb, T.A. (1980), Testosterone, status and mood in human males, Horm. Behav. 14:236-246.

McCarthy, R. & Southwick, C.H. (1979) Parental environment: effects of survival, growth and aggressive behaviors of two rodent species. Dev. Psychobiol. 12:269-279.

Meyer-Bahlburg, H.F.L. (1974), Aggression and androgens and the XYY syndrome. In:Sex Differences in Behavior, eds. R..C. Friedman, R.M. Richart & R.L. Vande Wiele. New York: John Wiley, pp.433-453.

Monti, P.M., Brown, W.A. & Corriveau, M.A. (1977), Testosterone and components of aggressive and sexual behavior in man. Am. J. Psychiatry, 134:692-694.

Mueller, E. & Silverman, N. (1989), Peer relations in maltreated children. In: Child Maltreatment: Theory and Research on the Causes and Consequences of Child Abuse and Neglect, eds. D. Cicchetti, & V. Carlson. New York: Cambridge University Press, pp. 529-578.

Mugford, R.A. & Nowell, N.W. (1972), Paternal stimulation during infancy: effects upon aggression and open-field performance of mice. Journal of Comparative and Physiological Psychology, 79:30-36.

Nielson, J. (1968), The XXY syndrome in a mental hospital. British Journal of Criminology, 8:186-203.

Owen, D.R. (1972), The 47XYY male: a review. Psychol. Bull. 79:209-233.

Palmour, R.M. (1983). Genetic models for the study of aggressive behavior. Prog. Neuropsychopharmacol. Biol. Psychiatry. 7:513-517.

Pasamanick, B. (1956). Pregnancy experience and the development of behavior disorders in children. Am. J. Psychiatry 112:613-617.

--(1961), Epidemiological investigations of some prenatal factors in the production of neuropsychiatric disorder. In: Comparative Epidemiology for Mental Disorders, eds. P.H. Hoch & J. Zubin. New York: Grune & Stratton, pp. 260-275.

Patterson, G.R. (1977). Accelerating stimuli for two classes of coercive behaviors. J. Abnorm. Child Psychol. 5:335-350.

--(1979), A performance theory for coercive family interaction. In: The Analysis of Social Interactions: Methods, Issues, and Illustrations, ed. R.B. Caines, Hillsdale, NJ: Laurence Erlbaum.

Pope, H.G. & Katz, D.L. (1987). Bodybuilder's psychosis. Lancer, 1:863.

--(1988), Affective and psychotic symptoms associated with anabolic steroid use. Am. J. Psychiatry. 145:487-490.

Provence, S. & Lipton, R. (1962). Infants in Institutions. New York: International Universities.

Rada, R.T., Laws, D.R., & Kellner, R. (1976) Plasma testosterone levels in the rapist. Psychosom. Med. 38:257-268.

Randrup, A. & Munkvad, I. (1966), DOPA and other naturally occurring substances as causes of stereotype and rage in rats. Acra. Psychiatr. Scand. 42[Suppl. 19]:193.

Reis, D.J. (1972). Central Neurotransmiters in Aggression, S.F. Frazier. Aggression, 52:119-147.

Reisert, L., Han, V., Lieth, E., Toran, A.D., Pilgrim, C. & Lauder, J. (1987), Sex, steroids promote neurite growth in mesencephalic tyrosine hydroxylase immunorcactive neutron in vitro. International JOurnal of Developmental Neuroscience, 5:91-98.

Rieder, C. & Cicchetti, D. (1989), An organizational perspective on cognitive control functioning and cognitive-affective balance in maltreated children. Developmental Psychology, 25:382-393.

Rose, R., Holaday, J. & Bernstein, I. (1971). Plasma testosterone dominance rank and aggressive behaviour in male rhesus monkeys. Nature, 231:366.

Rotton, J., Barry, T., Frey, J. & Soler, E. (1978), Air pollution and interpersonal attraction. J. Appl. Soc. Psychol. 8:57.

Rutter, M. (1970). Sex differences in rsponse to family stress. In: The Child and His Family, eds. E.J. Anthony & C. Kompernik. New York: Wiley.

Sahakian, B.J. (1981), The neurochemical basis of hyperactivity and aggression induced by social deprivation. In: Vulnerabilities to Delinquency. Ed. D.O. Lewis. New York: Spectrum. pp. 173-186.

Schiavo, R., Theilgaard, A., Owen, D. et al (1984), Sex, chromosome anomalies, hormones and aggressivity. Arch. Gen. Psychiatry, 4:93-99.

Seward, J.F. (1946), Aggressive behavior in the rat: IV Submission determine by conditioning, extinction and disuse. J. Comp. Psychol. 39:51.

Siegel, A. & Flynn, J.P. (1968), Differential effects of electrical stimulations and lesions of the hippocampus and adjacent regions upon attack behavior of cats. Brain Rev. 7:252-267.

Smetana, J.G. & Kelly, M. (1989), Social cognition in maltreated children. In: Child Maltreatment, eds. D. Chicchetti & V. Carlson. Cambridge: Cambridge U. Press. pp. 620-646.

Smith, D.E., King, M.D., Hoebel, B.G. (1970), Lateral hyppothalamic control of killing evidence for a cholinoceptive mechanism. Science 167:900-901.

Smith, M.L. & Simmel, B.C. (1977), Paternal effects on the development of social behavior in Mus Musculus In: Child Maltreatment Theory and Research on the Causes and Consequences of Child Abuse and Neglect. eds. D. Cicchetti & V. Carlson. New York: Cambridge University Press. pp.620-646.

Southwick, C.H. (1968). Effect of maternal environment on aggressive behavior of inbred mice. Communications in Behavioral Biology. 1:129-132.

Spitz, R.A. (1946), Anaclitic depression. Psychoanal. Study Child. 2:313-342.

Stott, D.H. (1973). Follow-up study from birth of the effects of prenatal stresses. Dev. Med. Child Neurol. 15:770-787.

--Latchford, S.A. (1976), Prenatal antecedents of child health, development and behavior. J. Am. Acad. Child Adolesc. Psychiatry. 15:161-191.

Telfer, M.A. (1968), Are some criminals born that way? Think, 34:24-28.

Tennant, F., Black, D.L. & Voy, R.O. (1988), Anabolic steroid dependence with opioid-type features. N. Engl. J. Med. 319-578.

Tupin, J.P. (1987). Psychopharmacology and aggression. In: Clinical Treatment of the Violent Person, ed. L.H. Roth. pp. 79-94.

Valzelli, L. (1974), 5-Hydroxytryptamine in aggressiveness. In: Advances in Biochemical Psycopharmacology, eds. E. Costa, G. Gessa & M. Sandler, New York: Raven.

Virkkunen, M., DeJong, J., Barko, J., Goodwin, F.K. & Linnolla, M. (1989), Relationship of psychobiological variables to recidivism in violent offenders and impulsive fire setters. Arch. Gen. Psychiatry 46:600-603.

vom Saal, F.S. (1979), Prenatal exposure to androgen influence morphology and aggressive behavior of male and female mice. Horm. Behav., 12:1.

---(1984), The intrauterine position phenomenon: effects on physiology, aggressive behavior and population dynamics in house mice. In: Biological Perspectives on Aggression, eds. K.J. Flannelly, R. J. Blanchard & D.C. Blanchard. New York: Liss. pp. 135-179.

Weiger, W.A. & Bear D.M. (1988), AN approach to the neurology of aggression. J. Psychiat. Res. 22:85-98.

Widom, C.S. (1989), The cycle of violence. Science 244:160-166.

Witkin, H.A., Mednick, S.A., Schuylsinger, F. et al (1976), Criminality in XYY and XXY men, Science 193:547-555.

Yesavage, J.A. (1983a), Bipolar Illness: correlates of dangerous impatient behavior. Br. J. Psychiatry 143:554-557.

---(1983b), Correlates of dangerous behavior by schizophrenics in hospital. J. Psychosom. Res. 18:225-231.