Impact News

Responding to Violence, Suicide, Psychosis and Trauma

Does Biology Play a Role in Domestic Violence?

TUCSON—Between 20% and 30% of all men and women in the US will be victims of domestic violence in their lifetime. Domestic violence accounts for 20% of all emergency department visits, 50% of police calls, and about 30% of murdered women. While considerable research into understanding the perpetrator’s mindset has focused on learned behaviors and psychosocial issues, comparatively little effort has been devoted to exploring possible biological causes of the problem, according to David George, MD.

“Most people look at domestic violence from a psychodynamic/psychosocial perspective,” said Dr. George, Section Chief of Clinical and Translational Studies at the National Institute on Alcohol Abuse and Alcoholism in Bethesda, Maryland. “These people believe that perpetrators feel inadequate and try to control other people by their behaviors or that they grew up in homes where they were exposed to violence, and, therefore, they’ve learned these patterns. I was particularly interested in the fact that there has been so little emphasis given to any biological understanding of what might be taking place.” Dr. George made his presentation at the 18th Annual Meeting of the American Neuropsychiatric Association.

The first step in determining whether biological abnormalities may lead to acts of domestic violence is to closely examine who the perpetrators are, according to Dr. George. The incidence of domestic violence is approximately equal in men and women, and about 70% of perpetrators abuse alcohol, he noted. Based on interviews with several hundred people who have committed acts of domestic violence, as well as their spouses and significant others, Dr. George has observed several recurring patterns. One of these patterns is that perpetrators are likely to have been in multiple fights during their childhood. “They are going to push their teachers,” noted Dr. George. “They fight with their siblings and with the kids down the street. As they grow older, most of them tend to limit their violence to the home and direct it toward their spouse or significant other.”

Perpetrators also have little insight into why they become violent, and most acts of domestic violence are impulsive, said Dr. George. “There are those with a predatory side, but I do not see it often. Alcohol plays an important role in domestic violence. Alcohol is a two-edged sword. Perpetrators are going to use alcohol to calm down, but often the alcohol contributes to the likelihood of violence.”

Typical behavioral symptoms in perpetrators include racing thoughts, supersensitivity to environmental stimuli, and mood swings that range from shutdown to flight, fight, and stalking. “I had one person tell me, ‘If you ever got in my mind, you would probably lock me up. You would think I was crazy.’ This is something that is going on inside of them,” said Dr. George. “Little things are going to set them off—spilled milk at the dinner table, dirty dishes that aren’t taken care of in the sink, the dinner that’s late. The most interesting thing was that they feel afraid at the time of the aggression. That was very difficult for me to comprehend, because so often we are working with large and aggressive perpetrators whose victims are smaller in stature. Fear just doesn’t look like it should be a significant factor.”


Dr. George has conducted a number of studies regarding domestic violence. One trial included perpetrators of domestic violence with alcohol dependence, nonviolent alcoholics, and healthy controls. The researchers found that violent alcoholics had a higher incidence of major depression, panic attacks, social phobia, obsessive-compulsive disorder, generalized anxiety, and certain personality disorders than did nonviolent alcoholics.

In a double-blind, placebo-controlled trial involving the administration of sodium lactate to participants, Dr. George and colleagues found that behavioral symptoms such as speech, breathing, facial grimacing, and motor activity in the arms and legs were much more accentuated in the perpetrators, as was their sense of fear, panic, and rage, compared with nonviolent controls. “These results were instrumental in changing my thinking about perpetrators of domestic violence,” commented Dr. George. “It moved me from seeing them as offensive individuals to seeing them as defensive individuals. This was extremely important to me, because it directed my attention to the neuropathways that have been shown in animals to mediate defensive aggression.”


Dr. George devised a basic model for understanding the psychopathology of perpetrators of domestic violence. “Perpetrators frequently misinterpret environmental stimuli, which gives rise to a perceived sense of threat,” he explained. “Sensory stimuli enter the thalamus, and from there are processed by both the cortex and the amygdala. The processing of the sensory stimuli in the amygdala is extremely fast and serves as an early warning system. The processing of the sensory stimuli in the cortex is going to be much slower and much more detailed than in the amygdala…. The cortex and the amygdala talk to each other. In certain situations, these sensory stimuli give rise to defensive behavior, autonomic arousal, and hypoalgesia…. If you talk to these people and ask them what it is like when they are hitting someone, they will tell you, ‘It feels like my hands and arms are like feathers. I have no feeling in my hands. I don’t feel as though I’m doing anything.’”

In formulating a theory for the etiology of domestic violence, Dr. George reasoned that threats trigger a conditioned fear response in perpetrators that is out of proportion to the stimulus, which may result in fear-induced aggression. “This misinterpretation arises from the abnormality in structures and pathways that mediate fear-induced aggression,” he said.

In a study using PET (18FDG) imaging to examine the neural structures and pathways involved in fear conditioning and fear-induced aggression, Dr. George’s group found that mean CMRglc in the right hypothalamus was significantly lower in perpetrators with alcohol dependence, compared with nonviolent alcoholics and healthy controls. “At rest, when you compare the activities in the left amygdala with various cortical and subcortical structures like the thalamus and cingulate, you see a strong correlation in the nonviolent alcoholics between these structures and the amygdala, whereas in the perpetrators, you had decreased correlations,” said Dr. George. “We are interpreting this to mean that the ability of the cortex to modulate the amygdala in these people is reduced. Similarly, we compared perpetrators with healthy controls. We found the same kind of finding here, decreased correlations [with the left and right amygdala]. And the nonviolent alcoholics had an increased correlation between the left thalamus and left posterior orbitofrontal cortex.”

Such findings may indicate different motivations to drink alcohol for nonviolent alcoholics and alcoholic perpetrators. “Basically, we arrived at two different possibilities,” Dr. George said. “The increased correlation found in nonviolent alcoholics maybe makes them more susceptible to environmental cues that trigger drinking. Whereas, I think alcoholic perpetrators are more prone, at least in the initial stages of the disease, to drink in order to decrease anxiety.”

In another study, Dr. George and colleagues performed lumbar puncture in the left lateral decubitus position in alcoholic perpetrators of domestic violence, nonalcoholic perpetrators, and healthy controls. The researchers found that the nonalcoholic violent group had lower 5-HIAA [5-hydroxyindoleacetic acid] concentrations than did the other two groups, which was “not particularly surprising, given the huge literature that’s out there saying that 5-HIAA is involved with impulsive types of aggression,” noted Dr. George. “It is unclear as to why the alcoholics didn’t have it. We then looked at testosterone, and there we found that [alcoholic perpetrators] did have higher levels of testosterone. So we have at least two neurotransmitter systems that theoretically could be involved, that could be modulating the way they process sensory information. We are looking at a number of other transmitter systems at this time.”


Dr. George’s current research is focusing on fMRI, genotyping, and potential treatments. To date, he emphasized, “Treatments for domestic violence are often ineffective.” In one ongoing trial, he has been comparing fluoxetine with placebo regarding their effect on measures of aggression, anxiety, and depression in those who commit acts of domestic violence. “What is really interesting is when you look at what serotonin does, it modulates sensory information,” noted Dr. George.

Dr. George believes that it is possible to piece together some of these findings to understand domestic violence on the basis of a biological pathway. “This is such a primitive pathway,” he commented. “Defensive aggression is present throughout the whole animal kingdom and promotes survival. With reduced cortical connection to the amygdala, perpetrators process sensory information very quickly. Based on fMRI studies, this processing of sensory information by the amygdala is out of the conscious awareness. I think that’s why therapy has been so ineffective in these individuals. They are responding so quickly to sensory information that they don’t even have time to think about it.”

Ultimately, Dr. George believes that further studies linking conditioned fear and fear avoidance with behaviors and psychiatric diagnoses will help change the way researchers and clinicians perceive and treat perpetrators of domestic violence.           

—Colby Stong

Suggested Reading
Fils-Aime ML, Eckardt MJ, George DT, et al. Early-onset alcoholics have lower cerebrospinal fluid 5-hydroxyindoleacetic acid levels than late-onset alcoholics. Arch Gen Psychiatry. 1996;53:211-216.
George DT, Phillips MJ, Doty L, et al. A model linking biology, behavior and psychiatric diagnoses in perpetrators of domestic violence. Med Hypotheses. 2006;67:345-353.
George DT, Umhau JC, Phillips MJ, et al. Serotonin, testosterone and alcohol in the etiology of domestic violence. Psychiatry Res. 2001;104:27-37.
Umhau JC, Petrulis SG, Diaz R, Rawlings R, George DT. Blood glucose is correlated with cerebrospinal fluid neurotransmitter metabolites. Neuroendocrinology. 2003;78:339-343.


Filed under: Violence, , , , , , , , ,