Aggression is overt or covert, often harmful, social interaction with the intention of inflicting damage or other unpleasantness upon another individual. It may occur either reactively or without provocation. In humans, frustration due to blocked goals can cause aggression. Human aggression can be classified into direct and indirect aggression; whilst the former is characterized by physical or verbal behavior intended to cause harm to someone, the latter is characterized by behavior intended to harm the social relations of an individual or group.^[1][2]

In definitions commonly used in the social sciences and behavioral sciences, aggression is an action or response by an individual that delivers something unpleasant to another person.^[3] Some definitions include that the individual must intend to harm another person.^[4] Predatory or defensive behavior between members of different species may not be considered aggression in the same sense.

Aggression can take a variety of forms, which may be expressed physically, or communicated verbally or non-verbally: including anti-predator aggression, defensive aggression (fear-induced), predatory aggression, dominance aggression, inter-male aggression, resident-intruder aggression, maternal aggression, species-specific aggression, sex-related aggression, territorial aggression, isolation-induced aggression, irritable aggression, and brain-stimulation-induced aggression (hypothalamus). There are two subtypes of human aggression: (1) controlled-instrumental subtype (purposeful or goal-oriented); and (2) reactive-impulsive subtype (often elicits uncontrollable actions that are inappropriate or undesirable). Aggression differs from what is commonly called assertiveness, although the terms are often used interchangeably among laypeople (as in phrases such as "an aggressive salesperson").^[5]

Aggression can have adaptive benefits or negative effects.

Aggressive behavior is an individual or collective social interaction that is a hostile behavior with the intention of inflicting damage or harm.^[1][2] Two broad categories of aggression are commonly distinguished. One includes affective (emotional) and hostile, reactive, or retaliatory aggression that is a response to provocation, and the other includes instrumental, goal-oriented or predatory, in which aggression is used as a mean to achieve a goal.^[6] An example of hostile aggression would be a person who punches someone who insulted him or her. An instrumental form of aggression would be armed robbery. Research on violence from a range of disciplines lend some support to a distinction between affective and predatory aggression.^[7] However, some researchers question the usefulness of a hostile versus instrumental distinction in humans, despite its ubiquity in research, because most real-life cases involve mixed motives and interacting causes.^[8]

A number of classifications and dimensions of aggression have been suggested.

These depend on such things as whether the aggression is verbal or physical; whether or not it involves relational aggression such as covert bullying and social manipulation;^[9] whether harm to others is intended or not; whether it is carried out actively or expressed passively; and whether the aggression is aimed directly or indirectly. Classification may also encompass aggression-related emotions (e.g. anger) and mental states (e.g. impulsivity, hostility).^[10] Aggression may occur in response to non-social as well as social factors, and can have a close relationship with stress coping style.^[11] Aggression may be displayed in order to intimidate.

The operative definition of aggression may be affected by moral or political views. Examples are the axiomatic moral view called the non-aggression principle and the political rules governing the behavior of one country toward another.^[12] Likewise in competitive sports, or in the workplace, some forms of aggression may be sanctioned and others not (see Workplace aggression).^[13] Aggressive behaviors are associated with adjustment problems and several psychopathological symptoms such as Antisocial Personality Disorder, Borderline Personality Disorder, and Intermittent Explosive Disorder.^[14]

Biological approaches conceptualize aggression as an internal energy released by external stimuli, a product of evolution through natural selection, part of genetics, a product of hormonal fluctuations.

Psychological approaches conceptualize aggression as a destructive instinct, a response to frustration, an affect excited by a negative stimulus, a result of observed learning of society and diversified reinforcement, a resultant of variables that affect personal and situational environments.^[15][16]

The term aggression comes from the Latin word aggressio, meaning attack. The Latin was itself a joining of ad- and gradi-, which meant step at. The first known use dates back to 1611, in the sense of an unprovoked attack.^[17] A psychological sense of "hostile or destructive behavior" dates back to a 1912 English translation of Sigmund Freud's writing.^[18] Alfred Adler theorized about an "aggressive drive" in 1908. Child raising experts began to refer to aggression, rather than anger, from the 1930s.^[19]

Ethologists study aggression as it relates to the interaction and evolution of animals in natural settings. In such settings aggression can involve bodily contact such as biting, hitting or pushing, but most conflicts are settled by threat displays and intimidating thrusts that cause no physical harm. This form of aggression may include the display of body size, antlers, claws or teeth; stereotyped signals including facial expressions; vocalizations such as bird song; the release of chemicals; and changes in coloration.^[20] The term agonistic behaviour is sometimes used to refer to these forms of behavior.

Most ethologists believe that aggression confers biological advantages.

Aggression may help an animal secure territory, including resources such as food and water. Aggression between males often occurs to secure mating opportunities, and results in selection of the healthier/more vigorous animal. Aggression may also occur for self-protection or to protect offspring.^[21] Aggression between groups of animals may also confer advantage; for example, hostile behavior may force a population of animals into a new territory, where the need to adapt to a new environment may lead to an increase in genetic flexibility.^[22]

The most apparent type of interspecific aggression is that observed in the interaction between a predator and its prey. However, according to many researchers, predation is not aggression. A cat does not hiss or arch its back when pursuing a rat, and the active areas in its hypothalamus resemble those that reflect hunger rather than those that reflect aggression.^[23] However, others refer to this behavior as predatory aggression, and point out cases that resemble hostile behavior, such as mouse-killing by rats.^[24] In aggressive mimicry a predator has the appearance of a harmless organism or object attractive to the prey; when the prey approaches, the predator attacks.

An animal defending against a predator may engage in either "fight or flight" or "tend and befriend" in response to predator attack or threat of attack, depending on its estimate of the predator's strength relative to its own. Alternative defenses include a range of antipredator adaptations, including alarm signals. An example of an alarm signal is nerol, a chemical which is found in the mandibular glands of Trigona fulviventris individuals.^[25] Release of nerol by T. fulviventris individuals in the nest has been shown to decrease the number of individuals leaving the nest by fifty percent, as well as increasing aggressive behaviors like biting.^[25] Alarm signals like nerol can also act as attraction signals; in T. fulviventris, individuals that have been captured by a predator may release nerol to attract nestmates, who will proceed to attack or bite the predator.^[25]

Aggression between groups is determined partly by willingness to fight, which depends on a number of factors including numerical advantage, distance from home territories, how often the groups encounter each other, competitive abilities, differences in body size, and whose territory is being invaded.^[26] Also, an individual is more likely to become aggressive if other aggressive group members are nearby.^[27] One particular phenomenon – the formation of coordinated coalitions that raid neighbouring territories to kill conspecifics – has only been documented in two species in the animal kingdom: 'common' chimpanzees and humans.^[28]

Aggression between conspecifics in a group typically involves access to resources and breeding opportunities.

One of its most common functions is to establish a dominance hierarchy. This occurs in many species by aggressive encounters between contending males when they are first together in a common environment.^[29] Usually the more aggressive animals become the more dominant.^[30][31] In test situations, most of the conspecific aggression ceases about 24 hours after the group of animals is brought together.^[29][32] Aggression has been defined from this viewpoint as "behavior which is intended to increase the social dominance of the organism relative to the dominance position of other organisms".^[33] Losing confrontations may be called social defeat, and winning or losing is associated with a range of practical and psychological consequences.^[34]

Conflicts between animals occur in many contexts, such as between potential mating partners, between parents and offspring, between siblings and between competitors for resources.

Group-living animals may dispute over the direction of travel or the allocation of time to joint activities.

Various factors limit the escalation of aggression, including communicative displays, conventions, and routines.

In addition, following aggressive incidents, various forms of conflict resolution have been observed in mammalian species, particularly in gregarious primates. These can mitigate or repair possible adverse consequences, especially for the recipient of aggression who may become vulnerable to attacks by other members of a group. Conciliatory acts vary by species and may involve specific gestures or simply more proximity and interaction between the individuals involved. However, conflicts over food are rarely followed by post conflict reunions, even though they are the most frequent type in foraging primates.^[35]

Other questions that have been considered in the study of primate aggression, including in humans, is how aggression affects the organization of a group, what costs are incurred by aggression, and why some primates avoid aggressive behavior.^[36] For example, bonobo chimpanzee groups are known for low levels of aggression within a partially matriarchal society. Captive animals including primates may show abnormal levels of social aggression and self-harm that are related to aspects of the physical or social environment; this depends on the species and individual factors such as gender, age and background (e.g. raised wild or captive).^[37]

Like many behaviors, aggression can be examined in terms of its ability to help an animal itself survive and reproduce, or alternatively to risk survival and reproduction.

This cost-benefit analysis can be looked at in terms of evolution. However, there are profound differences in the extent of acceptance of a biological or evolutionary basis for human aggression.^[38]

According to the male warrior hypothesis, intergroup aggression represents an opportunity for men to gain access to mates, territory, resources and increased status. As such, conflicts may have created selection evolutionary pressures for psychological mechanisms in men to initiate intergroup aggression.^[39][40]

Aggression can involve violence that may be adaptive under certain circumstances in terms of natural selection. This is most obviously the case in terms of attacking prey to obtain food, or in anti-predatory defense. It may also be the case in competition between members of the same species or subgroup, if the average reward (e.g. status, access to resources, protection of self or kin) outweighs average costs (e.g. injury, exclusion from the group, death). There are some hypotheses of specific adaptions for violence in humans under certain circumstances, including for homicide, but it is often unclear what behaviors may have been selected for and what may have been a byproduct, as in the case of collective violence.^{[41][42][43][44]}

Although aggressive encounters are ubiquitous in the animal kingdom, with often high stakes, most encounters that involve aggression may be resolved through posturing, or displaying and trial of strength.

Game theory is used to understand how such behaviors might spread by natural selection within a population, and potentially become 'Evolutionary Stable Strategies'. An initial model of resolution of conflicts is the hawk-dove game. Others include the Sequential assessment model and the Energetic war of attrition. These try to understand not just one-off encounters but protracted stand-offs, and mainly differ in the criteria by which an individual decides to give up rather than risk loss and harm in physical conflict (such as through estimates of resource holding potential).^[45]

Gender plays an important role in human aggression.^[46] There are multiple theories that seek to explain findings that males and females of the same species can have differing aggressive behaviors.

However, the conditions under which women and men differ in aggressiveness are not well understood or studied.^[47]

In general, sexual dimorphism can be attributed to greater intraspecific competition in one sex, either between rivals for access to mates and/or to be chosen by mates.^[48][49] This may stem from the other gender being constrained by providing greater parental investment, in terms of factors such as gamete production, gestation, lactation, or upbringing of young. Although there is much variation in species, generally the more physically aggressive sex is the male, particularly in mammals.^[50] In species where parental care by both sexes is required, there tends to be less of a difference. When the female can leave the male to care for the offspring, then females may be the larger and more physically aggressive. Competitiveness despite parental investment has also been observed in some species.^[51] A related factor is the rate at which males and females are able to mate again after producing offspring, and the basic principles of sexual selection are also influenced by ecological factors affecting the ways or extent to which one sex can compete for the other. The role of such factors in human evolution is controversial.

The pattern of male and female aggression is argued to be consistent with evolved sexually-selected behavioral differences,^[48] while alternative or complimentary views emphasize conventional social roles stemming from physical evolved differences.^[52] Aggression in women may have evolved to be, on average, less physically dangerous and more covert or indirect.^[53][54] However, there are critiques for using animal behavior to explain human behavior. Especially in the application of evolutionary explanations to contemporary human behavior, including differences between the genders.^[55]

According to the 2015 International Encyclopedia of the Social & Behavioral Sciences, sex differences in aggression is one of the most robust and oldest findings in psychology.^[56] Past meta-analyses in the encyclopedia found males regardless of age engaged in more physical and verbal aggression while small effect for females engaging in more indirect aggression such as rumor spreading or gossiping.^[56] It also found males tend to engage in more unprovoked aggression at higher frequency than females.^[56] This analysis also conforms with the Oxford Handbook of Evolutionary Psychology which reviewed past analysis which found men to use more verbal and physical aggression with the difference being greater in the physical type.^[57] There are more recent findings that show that differences in male and female aggression appear at about two years of age, though the differences in aggression are more consistent in middle-aged children and adolescence. Tremblay, Japel and Pérusse (1999) asserted that physically aggressive behaviors such as kicking, biting and hitting are age-typical expressions of innate and spontaneous reactions to biological drives such as anger, hunger, and affiliation.^[58] Girls' relational aggression, meaning non-physical or indirect, tends to increase after age two while physical aggression decreases. There was no significant difference in aggression between males and females before two years of age.^[59] A possible explanation for this could be that girls develop language skills more quickly than boys, and therefore have better ways of verbalizing their wants and needs. They are more likely to use communication when trying to retrieve a toy with the words "Ask nicely" or "Say please."^[60]

According to the journal of Aggressive Behaviour, an analysis across 9 countries found boys reported more in the use of physical aggression.^[59] At the same time no consistent sex differences emerged within relational aggression.^[59] It has been found that girls are more likely than boys to use reactive aggression and then retract, but boys are more likely to increase rather than to retract their aggression after their first reaction. Studies show girls' aggressive tactics included gossip, ostracism, breaking confidences, and criticism of a victim's clothing, appearance, or personality, whereas boys engage in aggression that involves a direct physical and/or verbal assault.^[61] This could be due to the fact that girls' frontal lobes develop earlier than boys, allowing them to self-restrain.^[60]

One factor that shows insignificant differences between male and female aggression is in sports.

In sports, the rate of aggression in both contact and non-contact sports is relatively equal.

Since the establishment of Title IX, female sports have increased in competitiveness and importance, which could contribute to the evening of aggression and the "need to win" attitude between both genders.

Among sex differences found in adult sports were that females have a higher scale of indirect hostility while men have a higher scale of assault.^[62] Another difference found is that men have up to 20 times higher levels of testosterone than women.

Some studies suggest that romantic involvement in adolescence decreases aggression in males and females, but decreases at a higher rate in females.

Females will seem more desirable to their mate if they fit in with society and females that are aggressive do not usually fit well in society, they can often be viewed as antisocial.

Female aggression is not considered the norm in society and going against the norm can sometimes prevent one from getting a mate.^[63] However, studies have shown that an increasing number of women are getting arrested for domestic violence charges.

In many states, women now account for a quarter to a third of all domestic violence arrests, up from less than 10 percent a decade ago.

The new statistics reflect a reality documented in research: women are perpetrators as well as victims of family violence.^[64] However, another equally possible explanation is a case of improved diagnostics: it has become more acceptable for men to report female domestic violence to the authorities while at the same time actual female domestic violence has not increased at all.

This can be the case when men have become less ashamed of reporting female violence against them, therefore an increasing number of women are arrested, although the actual number of violent women remains the same.

In addition, males in competitive sports are often advised by their coaches not to be in intimate relationships based on the premises that they become more docile and less aggressive during an athletic event.

The circumstances in which males and females experience aggression are also different.

A study showed that social anxiety and stress was positively correlated with aggression in males, meaning as stress and social anxiety increases so does aggression.

Furthermore, a male with higher social skills has a lower rate of aggressive behavior than a male with lower social skills.

In females, higher rates of aggression were only correlated with higher rates of stress.

Other than biological factors that contribute to aggression there are physical factors are well.^[65]

Regarding sexual dimorphism, humans fall into an intermediate group with moderate sex differences in body size but relatively large testes. This is a typical pattern of primates where several males and females live together in a group and the male faces an intermediate number of challenges from other males compared to exclusive polygyny and monogamy but frequent sperm competition.^[66]

Evolutionary psychology and sociobiology have also discussed and produced theories for some specific forms of male aggression such as sociobiological theories of rape and theories regarding the Cinderella effect. Another evolutionary theory explaining gender differences in aggression is the Male Warrior hypothesis, which explains that males have psychologically evolved for intergroup aggression in order to gain access to mates, resources, territory and status.^[39][40]

Many researchers focus on the brain to explain aggression.

Numerous circuits within both neocortical and subcortical structures play a central role in controlling aggressive behavior, depending on the species, and the exact role of pathways may vary depending on the type of trigger or intention.^[67][2]

In mammals, the hypothalamus and periaqueductal gray of the midbrain are critical areas, as shown in studies on cats, rats, and monkeys. These brain areas control the expression of both behavioral and autonomic components of aggression in these species, including vocalization. Electrical stimulation of the hypothalamus causes aggressive behavior^[68] and the hypothalamus has receptors that help determine aggression levels based on their interactions with serotonin and vasopressin.^[69] In rodents, activation of estrogen receptor-expressing neurons in the ventrolateral portion of the ventromedial hypothalamus (VMHvl) was found to be sufficient to initiate aggression in both males and females.^[70][71] Midbrain areas involved in aggression have direct connections with both the brainstem nuclei controlling these functions, and with structures such as the amygdala and prefrontal cortex.

Stimulation of the amygdala results in augmented aggressive behavior in hamsters,^[72][73] while lesions of an evolutionarily homologous area in the lizard greatly reduce competitive drive and aggression (Bauman et al. 2006).^[74] In rhesus monkeys, neonatal lesions in the amygdala or hippocampus results in reduced expression of social dominance, related to the regulation of aggression and fear.^[75] Several experiments in attack-primed Syrian golden hamsters, for example, support the claim of circuity within the amygdala being involved in control of aggression.^[73] The role of the amygdala is less clear in primates and appears to depend more on situational context, with lesions leading to increases in either social affiliatory or aggressive responses.

The broad area of the cortex known as the prefrontal cortex (PFC) is crucial for self-control and inhibition of impulses, including inhibition of aggression and emotions. Reduced activity of the prefrontal cortex, in particular its medial and orbitofrontal portions, has been associated with violent/antisocial aggression.^[24] In addition, reduced response inhibition has been found in violent offenders, compared to non-violent offenders.^[67]

The role of the chemicals in the brain, particularly neurotransmitters, in aggression has also been examined. This varies depending on the pathway, the context and other factors such as gender. A deficit in serotonin has been theorized to have a primary role in causing impulsivity and aggression. At least one epigenetic study supports this supposition.^[77] Nevertheless, low levels of serotonin transmission may explain a vulnerability to impulsiveness, potential aggression, and may have an effect through interactions with other neurochemical systems. These include dopamine systems which are generally associated with attention and motivation toward rewards, and operate at various levels. Norepinephrine, also known as noradrenaline, may influence aggression responses both directly and indirectly through the hormonal system, the sympathetic nervous system or the central nervous system (including the brain). It appears to have different effects depending on the type of triggering stimulus, for example social isolation/rank versus shock/chemical agitation which appears not to have a linear relationship with aggression. Similarly, GABA, although associated with inhibitory functions at many CNS synapses, sometimes shows a positive correlation with aggression, including when potentiated by alcohol.^[78][79]

The hormonal neuropeptides vasopressin and oxytocin play a key role in complex social behaviours in many mammals such as regulating attachment, social recognition, and aggression. Vasopressin has been implicated in male-typical social behaviors which includes aggression. Oxytocin may have a particular role in regulating female bonds with offspring and mates, including the use of protective aggression. Initial studies in humans suggest some similar effects.^[80][81]

In human, aggressive behavior has been associated with abnormalities in three principal regulatory systems in the body serotonin systems, catecholamine systems, and the hypothalamic–pituitary–adrenal axis. Abnormalities in these systems also are known to be induced by stress, either severe, acute stress or chronic low-grade stress^[82]

Early androgenization has an organizational effect on the developing brains of both males and females, making more neural circuits that control sexual behavior as well as intermale and interfemale aggression become more sensitive to testosterone.^[83] Thus, aggressive behavior tends to increase with testosterone.

There are noticeable sex differences in aggression.

Testosterone is present to a lesser extent in females, who may be more sensitive to its effects.

Animal studies have also indicated a link between incidents of aggression and the individual level of circulating testosterone.

However, results in relation to primates, particularly humans, are less clear cut and are at best only suggestive of a positive association in some contexts.^[24]

In humans, there is a seasonal variation in aggression associated with changes in testosterone.^[85] For example, in some primate species, such as rhesus monkeys and baboons, females are more likely to engage in fights around the time of ovulation as well as right before menstruation.^[83] If the results were the same in humans as they are in rhesus monkeys and baboons, then the increase in aggressive behaviors during ovulation is explained by the decline in estrogen levels. This makes normal testosterone levels more effective.^[86] Castrated mice and rats exhibit lower levels of aggression. Males castrated as neonates exhibit low levels of aggression even when given testosterone throughout their development.

The challenge hypothesis outlines the dynamic relationship between plasma testosterone levels and aggression in mating contexts in many species. It proposes that testosterone is linked to aggression when it is beneficial for reproduction, such as in mate guarding and preventing the encroachment of intrasexual rivals. The challenge hypothesis predicts that seasonal patterns in testosterone levels in a species are a function of mating system (monogamy versus polygyny), paternal care, and male-male aggression in seasonal breeders. This pattern between testosterone and aggression was first observed in seasonally breeding birds, such as the song sparrow, where testosterone levels rise modestly with the onset of the breeding season to support basic reproductive functions.^[87] The hypothesis has been subsequently expanded and modified to predict relationships between testosterone and aggression in other species. For example, chimpanzees, which are continuous breeders, show significantly raised testosterone levels and aggressive male-male interactions when receptive and fertile females are present.^[88] Currently, no research has specified a relationship between the modified challenge hypothesis and human behavior, or the human nature of concealed ovulation, although some suggest it may apply.^[85]

Another line of research has focused on the proximate effects of circulating testosterone on the nervous system, as mediated by local metabolism within the brain.

Testosterone can be metabolized to estradiol by the enzyme aromatase, or to dihydrotestosterone (DHT) by 5α-reductase.^[89]

Aromatase is highly expressed in regions involved in the regulation of aggressive behavior, such as the amygdala and hypothalamus.

In studies using genetic knockout techniques in inbred mice, male mice that lacked a functional aromatase enzyme displayed a marked reduction in aggression.

Long-term treatment with estradiol partially restored aggressive behavior, suggesting that the neural conversion of circulating testosterone to estradiol and its effect on estrogen receptors influences inter-male aggression. In addition, two different estrogen receptors, ERα and ERβ, have been identified as having the ability to exert different effects on aggression in mice. However, the effect of estradiol appears to vary depending on the strain of mouse, and in some strains it reduces aggression during long days (16 h of light), while during short days (8 h of light) estradiol rapidly increases aggression.^[89]

Another hypothesis is that testosterone influences brain areas that control behavioral reactions.

Studies in animal models indicate that aggression is affected by several interconnected cortical and subcortical structures within the so-called social behavior network. A study involving lesions and electrical-chemical stimulation in rodents and cats revealed that such a neural network consists of the medial amygdala, medial hypothalamus and periaqueductal grey (PAG), and it positively modulates reactive aggression.^[90] Moreover, a study done in human subjects showed that prefrontal-amygdala connectivity is modulated by endogenous testosterone during social emotional behavior.^[91]

In human studies, testosterone-aggression research has also focused on the role of the orbitofrontal cortex (OFC). This brain area is strongly associated with impulse control and self-regulation systems that integrate emotion, motivation, and cognition to guide context-appropriate behavior.^[92] Patients with localized lesions to the OFC engage in heightened reactive aggression.^[93] Aggressive behavior may be regulated by testosterone via reduced medial OFC engagement following social provocation.^[92] When measuring participants' salivary testosterone, higher levels can predict subsequent aggressive behavioral reactions to unfairness faced during a task. Moreover, brain scanning with fMRI shows reduced activity in the medial OFC during such reactions. Such findings may suggest that a specific brain region, the OFC, is a key factor in understanding reactive aggression.

Scientists have for a long time been interested in the relationship between testosterone and aggressive behavior.

In most species, males are more aggressive than females.

Castration of males usually has a pacifying effect on aggressive behavior in males. In humans, males engage in crime and especially violent crime more than females. The involvement in crime usually rises in the early teens to mid teens which happen at the same time as testosterone levels rise. Research on the relationship between testosterone and aggression is difficult since the only reliable measurement of brain testosterone is by a lumbar puncture which is not done for research purposes. Studies therefore have often instead used more unreliable measurements from blood or saliva.^[94]

The Handbook of Crime Correlates, a review of crime studies, states most studies support a link between adult criminality and testosterone although the relationship is modest if examined separately for each sex. However, nearly all studies of juvenile delinquency and testosterone are not significant. Most studies have also found testosterone to be associated with behaviors or personality traits linked with criminality such as antisocial behavior and alcoholism. Many studies have also been done on the relationship between more general aggressive behavior/feelings and testosterone. About half the studies have found a relationship and about half no relationship.^[94]

Studies of testosterone levels of male athletes before and after a competition revealed that testosterone levels rise shortly before their matches, as if in anticipation of the competition, and are dependent on the outcome of the event: testosterone levels of winners are high relative to those of losers.

No specific response of testosterone levels to competition was observed in female athletes, although a mood difference was noted.^[95] In addition, some experiments have failed to find a relationship between testosterone levels and aggression in humans.^[96][14][97]

The possible correlation between testosterone and aggression could explain the "roid rage" that can result from anabolic steroid use,^[98][99] although an effect of abnormally high levels of steroids does not prove an effect at physiological levels.

Dehydroepiandrosterone (DHEA) is the most abundant circulating androgen hormone and can be rapidly metabolized within target tissues into potent androgens and estrogens. Gonadal steroids generally regulate aggression during the breeding season, but non-gonadal steroids may regulate aggression during the non-breeding season. Castration of various species in the non-breeding season has no effect on territorial aggression. In several avian studies, circulating DHEA has been found to be elevated in birds during the non-breeding season. These data support the idea that non-breeding birds combine adrenal and/or gonadal DHEA synthesis with neural DHEA metabolism to maintain territorial behavior when gonadal testosterone secretion is low. Similar results have been found in studies involving different strains of rats, mice, and hamsters. DHEA levels also have been studied in humans and may play a role in human aggression. Circulating DHEAS (its sulfated ester) levels rise during adrenarche (≈7 years of age) while plasma testosterone levels are relatively low. This implies that aggression in pre-pubertal children with aggressive conduct disorder might be correlated with plasma DHEAS rather than plasma testosterone, suggesting an important link between DHEAS and human aggressive behavior.^[89]

Glucocorticoid hormones have an important role in regulating aggressive behavior. In adult rats, acute injections of corticosterone promote aggressive behavior and acute reduction of corticosterone decreases aggression; however, a chronic reduction of corticosterone levels can produce abnormally aggressive behavior. In addition, glucocorticoids affect development of aggression and establishment of social hierarchies. Adult mice with low baseline levels of corticosterone are more likely to become dominant than are mice with high baseline corticosterone levels.^[89]

Glucocorticoids are released by the hypothalamic pituitary adrenal (HPA) axis in response to stress, of which cortisol is the most prominent in humans. Results in adults suggest that reduced levels of cortisol, linked to lower fear or a reduced stress response, can be associated with more aggression. However, it may be that proactive aggression is associated with low cortisol levels while reactive aggression may be accompanied by elevated levels. Differences in assessments of cortisol may also explain a diversity of results, particularly in children.^[24]

The HPA axis is related to the general fight-or-flight response or acute stress reaction, and the role of catecholamines such as epinephrine, popularly known as adrenaline.

In many animals, aggression can be linked to pheromones released between conspecifics. In mice, major urinary proteins (Mups) have been demonstrated to promote innate aggressive behavior in males,^[100][101] and can be mediated by neuromodulatory systems.^[102] Mups activate olfactory sensory neurons in the vomeronasal organ (VNO), a subsystem of the nose known to detect pheromones via specific sensory receptors, of mice^[101] and rats.^[103] Pheremones have also been identified in fruit flies, detected by neurons in the antenna, that send a message to the brain eliciting aggression; it has been noted that aggression pheremones have not been identified in humans.^[104]

In general, differences in a continuous phenotype such as aggression are likely to result from the action of a large number of genes each of small effect, which interact with each other and the environment through development and life.

In a non-mammalian example of genes related to aggression, the fruitless gene in fruit flies is a critical determinant of certain sexually dimorphic behaviors, and its artificial alteration can result in a reversal of stereotypically male and female patterns of aggression in fighting. However, in what was thought to be a relatively clear case, inherent complexities have been reported in deciphering the connections between interacting genes in an environmental context and a social phenotype involving multiple behavioral and sensory interactions with another organism.^[105]

In mice, candidate genes for differentiating aggression between the sexes are the Sry (sex determining region Y) gene, located on the Y chromosome and the Sts (steroid sulfatase) gene.

The Sts gene encodes the steroid sulfatase enzyme, which is pivotal in the regulation of neurosteroid biosynthesis.

It is expressed in both sexes, is correlated with levels of aggression among male mice, and increases dramatically in females after parturition and during lactation, corresponding to the onset of maternal aggression.^[73] At least one study has found a possible epigenetic signature (i.e. decreased methylation at a specific CpG site on the promoter region) of the serotonin receptor 5-HT3a that is associated with maternal aggression among human subjects.^[77]

Mice with experimentally elevated sensitivity to oxidative stress (through inhibition of copper-zinc superoxide dismutase, SOD1 activity) were tested for aggressive behavior.^[106] Males completely deficient in SOD1 were found to be more aggressive than both wild-type males and males that express 50% of this antioxidant enzyme. They were also faster to attack another male. The causal connection between SOD1 deficiency and increased aggression is not yet understood.

In humans, there is good evidence that the basic human neural architecture underpinning the potential for flexible aggressive responses is influenced by genes as well as environment.

In terms of variation between individual people, more than 100 twin and adoption studies have been conducted in recent decades examining the genetic basis of aggressive behavior and related constructs such as conduct disorders. According to a meta-analysis published in 2002, approximately 40% of variation between individuals is explained by differences in genes, and 60% by differences in environment (mainly non-shared environmental influences rather than those that would be shared by being raised together). However, such studies have depended on self-report or observation by others including parents, which complicates interpretation of the results. The few laboratory-based analyses have not found significant amounts of individual variation in aggression explicable by genetic variation in the human population. Furthermore, linkage and association studies that seek to identify specific genes, for example that influence neurotransmitter or hormone levels, have generally resulted in contradictory findings characterized by failed attempts at replication. One possible factor is an allele (variant) of the MAO-A gene which, in interaction with certain life events such as childhood maltreatment (which may show a main effect on its own), can influence development of brain regions such as the amygdala and as a result some types of behavioral response may be more likely. The generally unclear picture has been compared to equally difficult findings obtained in regard to other complex behavioral phenotypes.^[24][108] For example, both 7R and 5R, ADHD-linked VNTR alleles of dopamine receptor D4 gene are directly associated with the incidence of proactive aggression in the men with no history of ADHD.^[109]

Humans share aspects of aggression with non-human animals, and have specific aspects and complexity related to factors such as genetics, early development, social learning and flexibility, culture and morals. Konrad Lorenz stated in his 1963 classic, On Aggression, that human behavior is shaped by four main, survival-seeking animal drives. Taken together, these drives—hunger, fear, reproduction, and aggression—achieve natural selection.^[110] E. O. Wilson elaborated in On Human Nature that aggression is, typically, a means of gaining control over resources. Aggression is, thus, aggravated during times when high population densities generate resource shortages.^[111] According to Richard Leakey and his colleagues, aggression in humans has also increased by becoming more interested in ownership and by defending his or her property.^[112] However, UNESCO adopted the Seville Statement of Violence in 1989 that refuted claims, by evolutionary scientists, that genetics by itself was the sole cause of aggression.^[113][114]

Social and cultural aspects may significantly interfere with the distinct expression of aggressiveness.

For example, a high population density, when associated with a decrease of available resources, might be a significant intervening variable for the occurrence of violent acts.^[115]

Culture is one factor that plays a role in aggression.

Tribal or band societies existing before or outside of modern states have sometimes been depicted as peaceful 'noble savages'. The ǃKung people were described as 'The Harmless People' in a popular work by Elizabeth Marshall Thomas in 1958,^[116] while Lawrence Keeley's 1996 War Before Civilization suggested that regular warfare without modern technology was conducted by most groups throughout human history, including most Native American tribes.^[117] Studies of hunter-gatherers show a range of different societies. In general, aggression, conflict and violence sometimes occur, but direct confrontation is generally avoided and conflict is socially managed by a variety of verbal and non-verbal methods. Different rates of aggression or violence, currently or in the past, within or between groups, have been linked to the structuring of societies and environmental conditions influencing factors such as resource or property acquisition, land and subsistence techniques, and population change.^[118]

American psychologist Peter Gray hypothesizes that band hunter-gatherer societies are able to reduce aggression while maintaining relatively peaceful, egalitarian relations between members through various methods, such as fostering a playful spirit in all areas of life, the use of humor to counter the tendency of any one person to dominate the group, and non-coercive or "indulgent" child-rearing practices. Gray likens hunter-gatherer bands to social play groups, while stressing that such play is not frivolous or even easy at all times.^[119] According to Gray, "Social play—that is, play involving more than one player—is necessarily egalitarian. It always requires a suspension of aggression and dominance along with a heightened sensitivity to the needs and desires of the other players".^[120]

Joan Durrant at the University of Manitoba writes that a number of studies have found physical punishment to be associated with "higher levels of aggression against parents, siblings, peers and spouses", even when controlling for other factors.^[121] According to Elizabeth Gershoff at the University of Texas at Austin, the more that children are physically punished, the more likely they are as adults to act violently towards family members, including intimate partners.^[122] In countries where physical punishment of children is perceived as being more culturally accepted, it is less strongly associated with increased aggression; however, physical punishment has been found to predict some increase in child aggression regardless of culture.^[123] While these associations do not prove causality, a number of longitudinal studies suggest that the experience of physical punishment has a direct causal effect on later aggressive behaviors.^[121] In examining several longitudinal studies that investigated the path from disciplinary spanking to aggression in children from preschool age through adolescence, Gershoff concluded: "Spanking consistently predicted increases in children's aggression over time, regardless of how aggressive children were when the spanking occurred".^[124] similar results were found by Catherine Taylor at Tulane University in 2010.^[125] Family violence researcher Murray A. Straus argues, "There are many reasons this evidence has been ignored. One of the most important is the belief that spanking is more effective than nonviolent discipline and is, therefore, sometimes necessary, despite the risk of harmful side effects".^[126]

Analyzing aggression culturally or politically is complicated by the fact that the label 'aggressive' can itself be used as a way of asserting a judgement from a particular point of view. Whether a coercive or violent method of social control is perceived as aggression – or as legitimate versus illegitimate aggression – depends on the position of the relevant parties in relation to the social order of their culture. This in turn can relate to factors such as: norms for coordinating actions and dividing resources; what is considered self-defense or provocation; attitudes towards 'outsiders', attitudes towards specific groups such as women, the disabled or the lower status; the availability of alternative conflict resolution strategies; trade interdependence and collective security pacts; fears and impulses; and ultimate goals regarding material and social outcomes.^[115]

Cross-cultural research has found differences in attitudes towards aggression in different cultures. In one questionnaire study of university students, in addition to men overall justifying some types of aggression more than women, United States respondents justified defensive physical aggression more readily than Japanese or Spanish respondents, whereas Japanese students preferred direct verbal aggression (but not indirect) more than their American and Spanish counterparts.^[127] Within American culture, southern men were shown in a study on university students to be more affected and to respond more aggressively than northerners when randomly insulted after being bumped into, which was theoretically related to a traditional culture of honor in the Southern United States, or "saving face".^[128] Other cultural themes sometimes applied to the study of aggression include individualistic versus collectivist styles, which may relate, for example, to whether disputes are responded to with open competition or by accommodating and avoiding conflicts. In a study including 62 countries school principals reported aggressive student behavior more often the more individualist, and hence less collectivist, their country's culture.^[129] Other comparisons made in relation to aggression or war include democratic versus authoritarian political systems and egalitarian versus stratified societies.^[115] The economic system known as capitalism has been viewed by some as reliant on the leveraging of human competitiveness and aggression in pursuit of resources and trade, which has been considered in both positive and negative terms.^[130] Attitudes about the social acceptability of particular acts or targets of aggression are also important factors. This can be highly controversial, as for example in disputes between religions or nation states, for example in regard to the Arab–Israeli conflict.^[131][132]

Some scholars believe that behaviors like aggression may be partially learned by watching and imitating people's behavior, while other researchers have concluded that the media may have some small effects on aggression.^[133] There is also research questioning this view.^[134] For instance, a long-term outcome study of youth found no long-term relationship between playing violent video games and youth violence or bullying.^[135] One study suggested there is a smaller effect of violent video games on aggression than has been found with television violence on aggression. This effect is positively associated with type of game violence and negatively associated to time spent playing the games.^[136] The author concluded that insufficient evidence exists to link video game violence with aggression. However, another study suggested links to aggressive behavior.^[137]

According to philosopher and neuroscientist Nayef Al-Rodhan, "fear(survival)-induced pre-emptive aggression" is a human reaction to injustices that are perceived to threaten survival. It is often the root of the unthinkable brutality and injustice perpetuated by human beings. It may occur at any time, even in situations that appear to be calm and under control. Where there is injustice that is perceived as posing a threat to survival, "fear(survival)-induced pre-emptive aggression" will result in individuals taking whatever action necessary to be free from that threat.

Nayef Al-Rodhan argues that humans' strong tendency towards "fear(survival)-induced pre-emptive aggression" means that situations of anarchy or near anarchy should be prevented at all costs. This is because anarchy provokes fear, which in turn results in aggression, brutality, and injustice. Even in non-anarchic situations, survival instincts and fear can be very powerful forces, and they may be incited instantaneously. "Fear(survival)-induced pre-emptive aggression" is one of the key factors that may push naturally amoral humans to behave in immoral ways.^[138] Knowing this, Al-Rodhan maintains that we must prepare for the circumstances that may arise from humans' aggressive behavior. According to Al-Rodhan, the risk of this aggression and its ensuing brutality should be minimized through confidence-building measures and policies that promote inclusiveness and prevent anarchy.^[139]

The frequency of physical aggression in humans peaks at around 2–3 years of age.

It then declines gradually on average.^[140][141] These observations suggest that physical aggression is not only a learned behavior but that development provides opportunities for the learning and biological development of self-regulation. However, a small subset of children fail to acquire all the necessary self-regulatory abilities and tend to show atypical levels of physical aggression across development. These may be at risk for later violent behavior or, conversely, lack of aggression that may be considered necessary within society. Some findings suggest that early aggression does not necessarily lead to aggression later on, however, although the course through early childhood is an important predictor of outcomes in middle childhood. In addition, physical aggression that continues is likely occurring in the context of family adversity, including socioeconomic factors. Moreover, 'opposition' and 'status violations' in childhood appear to be more strongly linked to social problems in adulthood than simply aggressive antisocial behavior.^[142][143] Social learning through interactions in early childhood has been seen as a building block for levels of aggression which play a crucial role in the development of peer relationships in middle childhood.^[144] Overall, an interplay of biological, social and environmental factors can be considered.^[145] Some research indicates that changes in the weather can increase the likelihood of children exhibiting deviant behavior.^[146]

The Bobo doll experiment was conducted by Albert Bandura in 1961. In this work, Bandura found that children exposed to an aggressive adult model acted more aggressively than those who were exposed to a nonaggressive adult model. This experiment suggests that anyone who comes in contact with and interacts with children can affect the way they react and handle situations.^[147]

Gender is a factor that plays a role in both human and animal aggression.

Males are historically believed to be generally more physically aggressive than females from an early age,^[150][151] and men commit the vast majority of murders (Buss 2005). This is one of the most robust and reliable behavioral sex differences, and it has been found across many different age groups and cultures. However, some empirical studies have found the discrepancy in male and female aggression to be more pronounced in childhood and the gender difference in adults to be modest when studied in an experimental context.^[47] Still, there is evidence that males are quicker to aggression (Frey et al. 2003) and more likely than females to express their aggression physically.^[152] When considering indirect forms of non-violent aggression, such as relational aggression and social rejection, some scientists argue that females can be quite aggressive, although female aggression is rarely expressed physically.^{[153][154][155]} An exception is intimate partner violence that occurs among couples who are engaged, married, or in some other form of intimate relationship.

Although females are less likely than males to initiate physical violence, they can express aggression by using a variety of non-physical means.

Exactly which method women use to express aggression is something that varies from culture to culture.

On Bellona Island, a culture based on male dominance and physical violence, women tend to get into conflicts with other women more frequently than with men. When in conflict with males, instead of using physical means, they make up songs mocking the man, which spread across the island and humiliate him. If a woman wanted to kill a man, she would either convince her male relatives to kill him or hire an assassin. Although these two methods involve physical violence, both are forms of indirect aggression, since the aggressor herself avoids getting directly involved or putting herself in immediate physical danger.^[156]

See also the sections on testosterone and evolutionary explanations for gender differences above.

There has been some links between those prone to violence and their alcohol use.

Those who are prone to violence and use alcohol are more likely to carry out violent acts.^[157] Alcohol impairs judgment, making people much less cautious than they usually are (MacDonald et al. 1996).

It also disrupts the way information is processed (Bushman 1993, 1997; Bushman & Cooper 1990).

Pain and discomfort also increase aggression.

Even the simple act of placing one's hands in hot water can cause an aggressive response.

Hot temperatures have been implicated as a factor in a number of studies.

One study completed in the midst of the civil rights movement found that riots were more likely on hotter days than cooler ones (Carlsmith & Anderson 1979).

Students were found to be more aggressive and irritable after taking a test in a hot classroom (Anderson et al. 1996, Rule, et al. 1987).

Drivers in cars without air conditioning were also found to be more likely to honk their horns (Kenrick & MacFarlane 1986), which is used as a measure of aggression and has shown links to other factors such as generic symbols of aggression or the visibility of other drivers.^[158]

Frustration is another major cause of aggression.

The Frustration aggression theory states that aggression increases if a person feels that he or she is being blocked from achieving a goal (Aronson et al. 2005). One study found that the closeness to the goal makes a difference. The study examined people waiting in line and concluded that the 2nd person was more aggressive than the 12th one when someone cut in line (Harris 1974). Unexpected frustration may be another factor. In a separate study to demonstrate how unexpected frustration leads to increased aggression, Kulik & Brown (1979) selected a group of students as volunteers to make calls for charity donations. One group was told that the people they would call would be generous and the collection would be very successful. The other group was given no expectations. The group that expected success was more upset when no one was pledging than the group who did not expect success (everyone actually had horrible success). This research suggests that when an expectation does not materialize (successful collections), unexpected frustration arises which increases aggression.

There is some evidence to suggest that the presence of violent objects such as a gun can trigger aggression.

In a study done by Leonard Berkowitz and Anthony Le Page (1967), college students were made angry and then left in the presence of a gun or badminton racket. They were then led to believe they were delivering electric shocks to another student, as in the Milgram experiment. Those who had been in the presence of the gun administered more shocks. It is possible that a violence-related stimulus increases the likelihood of aggressive cognitions by activating the semantic network.

A new proposal links military experience to anger and aggression, developing aggressive reactions and investigating these effects on those possessing the traits of a serial killer.

Castle and Hensley state, "The military provides the social context where servicemen learn aggression, violence, and murder."^[159] Post-traumatic stress disorder (PTSD) is also a serious issue in the military, also believed to sometimes lead to aggression in soldiers who are suffering from what they witnessed in battle. They come back to the civilian world and may still be haunted by flashbacks and nightmares, causing severe stress. In addition, it has been claimed that in the rare minority who are claimed to be inclined toward serial killing, violent impulses may be reinforced and refined in war, possibly creating more effective murderers.

Some recent scholarship has questioned traditional psychological conceptualizations of aggression as universally negative.^[33] Most traditional psychological definitions of aggression focus on the harm to the recipient of the aggression, implying this is the intent of the aggressor; however this may not always be the case.^[160] From this alternate view, although the recipient may or may not be harmed, the perceived intent is to increase the status of the aggressor, not necessarily to harm the recipient.^[161] Such scholars contend that traditional definitions of aggression have no validity.

From this view, rather than concepts such as assertiveness, aggression, violence and criminal violence existing as distinct constructs, they exist instead along a continuum with moderate levels of aggression being most adaptive.^[33] Such scholars do not consider this a trivial difference, noting that many traditional researchers' aggression measurements may measure outcomes lower down in the continuum, at levels which are adaptive, yet they generalize their findings to non-adaptive levels of aggression, thus losing precision.^[162]