The neuroscience of aggression
Lischinsky and Lin published a paper last year (2020) in Nature Neuroscience reviewing what we know (and don’t know) about the neural mechanisms behind aggression. A perfect paper from someone who is not in the field to learn about it from the very best.
The first thing that kind of surprised me is the definition of aggression. They define it as “hostile behaviors toward other animals of the same species”. So, we wouldn’t use the scientific term aggression for humans to attack other animals. What would we call it though? Later, in a summary box, they say that suicide is an extreme form of aggression, thus including in its definition hostile behavior toward itself, not only others. Maybe the definition is not so strict, then?
Anyway, the paper describes what seems to be a standard model in this kind of research, General Aggression Model (GAM). Several factors may increase your aggressive arousal and, once it’s over a given threshold, it triggers aggressive behavior.
Interestingly, they say that in humans, aggression-provoking stimulus occurs through several senses, and in animals is usually just one. Mice, for instance, can’t be aggressive if they lack a functional nose. This reminds me of what Bolsonaro does almost every day and that the press seems to be incapable of reporting correctly. He generates a lot of aggression-provoking stimuli, but not the stereotypical one. It’s the tone of speech, or sometimes a gesture, or even behavior (like not using masks). The speech, however, can be more modulated and contradict everything else. So he can speak about not being allowed to fight the pandemic, while not using masks or going into a crowded place. It seems to me also that the so-called passive-aggressive behavior involves this kind of use of multiple stimuli in conflicted ways (one signaling passivity or peace, and another one aggression).
Another interesting topic is how learning aggression happens. They explain that we can distinguish between innate and learned attack behaviors. Mice typically have only innate attack behavior, such as biting. However, they can learn arbitrary motor behaviors, such as lever pressing, which can be used by the researcher to generate an artificial situation in which a learned behavior is an attack. Mice can then learn to seek aggression, instead of only reacting to stimulus. This allows researchers to distinguish between reactive and proactive behavior. I don really understand the following, but they say that, since mice can learn to perform arbitrary actions that allow an opportunity of attack, they develop a preference for the cues that are associated with winning. So far so good. Then, they say that “like other innate social behaviors, such as sexual behaviors and maternal behaviors, the successful execution of aggressive behaviors is intrinsically rewarding and can serve as an unconditional stimulus for associative learning” (p. 6). Why does this mean that it is intrinsically rewarding?
The nuts and bolts of how this occurs, in which winner behavior is rewarded, are very interesting as well. If what would be a neutral stimulus (say, context) is paired [what does it mean for a stimulus to be paired? MG] with winning-induced dopamine release, the synapses involved in the neutral stimulus can be strengthened, which means that the context can be preferred in the next encounter. What do I understand from this? Learning is aided when you win when learning is happening. This makes me wonder how this happens in humans in the context of, say, politics. When a candidate runs for office, winning will strengthened several behaviors, including maybe being aggressive, is aggressive was associated with winning. A huge leap, of course, but make one wonder.
What causes aggression? They present evidence that when hunger, animals are more aggressive. Circadian rhythm is also associated with different levels of aggression, as well as reproductive state. These findings remember me about that famous finding that hunger judges were harsher in their sentencing, which was later found to not replicate. I thought that the proposed mechanism was kind of bogus, but it seems that the mechanism about hunger and aggression is more solid than I thought. Of course, translating to a complex behavior such as sentencing is hard and full of noises, but it makes me wonder if the original study is right, at least in the direction of the effect, if not in magnitude.
One important reason that may explain why all things causing aggression to not be expressed in aggressive behavior is because we can exercise control, mainly through the medial prefrontal cortex (mPFC) and lateral septum (LS). There is plenty of evidence for this for the role of the PFC in modulating aggression in humans (but not so much for LS). I mention here some of the evidence they cite for the role of PFC. Humans with injury in the frontal brain show a higher level of aggression; patients with psychiatric disorder with a history of violence show smaller PFC; stimulation of PFC decreases self-reported aggressiveness in imprisoned violent offenders. A lot of people talk about how Brazil is a very violent and authoritarian society. Would it be revealed in some measure of PFC or mPFC in comparison with other countries? Since we still know little of how the PFC controls aggression, it will be hard to find enough signal among noise on such data, if it is available at all.
The paper has a lot more information, especially about receptors, more specific parts of the brain, the difference in mice, birds, and rodents in general, among humans and other animals etc. I focused on a higher level summary, based on what interests me more. I recommend the reading to everyone who wants to know more.