A new study published in The Journal of Neuroscience has found that aggression can spread socially among mice, but only when there is a pre-existing bond between the observer and the aggressor. Researchers from Southern Illinois University School of Medicine have identified a brain region involved in this process and demonstrated that its activity is both necessary and sufficient for aggression learned through observation. These findings suggest a biological pathway by which violence may propagate within close social groups.
Aggression is not just a behavior individuals act out themselves. It can also be something they absorb by watching others. Classic work in human psychology showed that children can become more violent after watching aggressive adults. Later studies linked repeated exposure to aggression in peers or family to higher risks of criminal behavior. Similar effects have been seen in animals: mice tend to become more aggressive after watching other mice fight.
But scientists still do not fully understand how this process works. What are the brain mechanisms that allow a mouse — or potentially a person — to learn aggression by watching someone else? And are there specific social or environmental factors that make such learning more likely? Some research points to the importance of familiarity.
For example, rats tend to copy food choices only from cage mates, not from unfamiliar animals. In humans and animals alike, emotional contagion, such as fear or stress, often spreads more easily between individuals who know each other. Researchers in this study wanted to know whether familiarity also influences the way aggression is socially transmitted.
“Violence often spreads within families and peer groups, but the biological mechanisms behind this social transmission have been unknown,” said study author Jacob Nordman, an assistant professor of biomedical sciences at the Southern Illinois University School of Medicine.
“We wanted to understand how simply witnessing aggression could make someone more likely to act aggressively later. Earlier studies hinted that the brain’s social circuits might be sensitive to familiarity — whether the aggressor is known or a stranger — but no one had directly tested this in the context of aggression. Our goal was to identify both the behavioral and neural mechanisms that explain how exposure to familiar aggression can promote violence.”
The researchers focused on a specific part of the brain called the medial amygdala. This region processes social and sensory cues and plays a known role in aggressive behavior. In earlier work, the same research team showed that excitatory neurons in a posterior part of the medial amygdala can temporarily boost aggression after a fight. If these neurons are also activated while witnessing aggression, they might be involved in observational learning.
To examine this, the researchers developed a two-phase behavioral test in mice. In the first phase, one mouse, called the “witness,” was placed on one side of a transparent barrier. On the other side, a “demonstrator” mouse — either a familiar cage mate or a stranger — was allowed to interact with a third mouse, called the “intruder.” These interactions could include aggressive attacks. The witness mouse could see and hear everything but could not physically intervene. Thirty minutes later, in the second phase, the witness mouse was itself paired with a new intruder. Researchers measured whether the witness became aggressive during this encounter.
The researchers found that only witnesses who had observed a familiar demonstrator attack became more aggressive themselves. Witnesses who observed an unfamiliar mouse attacking showed no increase in aggression. These effects were specific to aggression and did not generalize to other types of social behaviors.
“We expected familiarity to matter, but the effect was striking — a single brief exposure to a familiar aggressor completely changed the observer’s behavior. That such a simple variable could ‘gate’ whether aggression spreads highlights how deeply social context shapes brain function.”
Next, the team used a method called fiber photometry to record real-time neural activity in the medial amygdala of the witnesses during the observation phase. They found that this brain region was significantly more active when a mouse watched a familiar demonstrator attack, compared to when it watched an unfamiliar one. This activity was absent in control animals that did not have the calcium-sensitive fluorescent protein used to detect neuron firing. Activity also increased slightly during non-aggressive social behavior — but only when that behavior came from a familiar demonstrator who had already shown aggression. This suggests the brain was interpreting even non-aggressive signals as potentially threatening in that context.
To see whether this brain activity played a functional role, the researchers then used chemogenetics to silence these neurons during the observation phase. Mice with silenced medial amygdala neurons were much less likely to become aggressive later, even after watching a familiar demonstrator attack. The same result came from optogenetic inhibition, which uses light to suppress neural activity. On the other hand, activating these neurons during the observation of an unfamiliar aggressor made the witness mouse more likely to attack later — even though unfamiliar demonstrators typically do not trigger this effect.
Together, these results suggest that the medial amygdala is not only activated during observed aggression, but its activation is required for the observer to internalize and later act out that aggression.
“Aggression can be ‘contagious,’ but only within familiar social circles. In mice, witnessing a familiar cage-mate fight was enough to make them more aggressive later — while watching a stranger had no effect. This was controlled by neurons in a region of the brain called the medial amygdala, which encodes social identity. These results suggest that violence tends to spread most easily within close social networks, offering insight into why cycles of aggression and abuse are often confined to families or peer groups.”
As with all animal studies, there are limits to how directly these findings can be applied to humans. While mice offer a useful model for studying neural circuits, their social systems and cognitive processes differ from ours. One surprising result was that female mice showed no sign of this type of socially transmitted aggression under any conditions. This could reflect true sex differences in aggression pathways, or it may be due to the specific strain of mice used. Future studies might examine other mouse strains or look more closely at hormonal cycles to clarify whether females can also learn aggression through observation.
“We’re now investigating how repeated exposure to familiar aggression changes the wiring of the medial amygdala and connected circuits. Additionally, we are very interested in why females seem impervious to observationally learned aggression. Ultimately, we hope this work will reveal neural targets that could be leveraged to reduce pathological aggression or the long-term effects of violent exposure.”
“The study doesn’t mean violence is inevitable or that watching aggression automatically makes one violent. Rather, it shows that social bonds can amplify how strongly we internalize others’ aggressive behavior. Understanding this mechanism could help design interventions that disrupt these cycles.”
The study, “Familiarity Gates Socially Transmitted Aggression via the Medial Amygdala,” was authored by Magdalene P. Adjei, Elana Qasem, Sophia Aaflaq, Jessica T. Jacobs, Savannah Skinner, Fletcher Summa, Claudia Spotanski, Rylee Thompson, Mikaela L. Aholt, Taylor Lineberry, and Jacob C. Nordman.