Measuring rewarding aspects of aggression

In his most famous book, Walden, the American author and naturalist Henry David Thoreau wrote that “many men go fishing all of their lives, without knowing it is not fish they are after”. Thus, one of the difficulties encountered when studying behaviour is to understand their underlying motivation. What makes some animals aggressive and other not when faced with the same situation? How does the brain process stimuli to generate an appropriate behavioural response?

A recent study by Golden and colleagues (Golden et al., 2016) has investigated this question in mice. They combined the resident-intruder assay (a rodent aggression test) with a condition place preference (CPP) test for reward behaviour. The resident-intruder test measures the response of a mouse towards an intruder in its home cage. In this case, male CD-1 mice were allowed to interact with a subordinate C57BL/BJ intruder. Aggressive contact was recorded as the latency for CB-1 mice to attack. Interestingly, about 70% of CD-1 mice were aggressive in this setup (AGGs) and 30% were non-aggressive (NONs). The CPP pairs a neutral stimulus (in this case one side of the home cage) with a conditioned stimulus – the intruder mouse. If the conditioned stimulus is rewarding then test animals spend more time on side of the cage where it was encountered. Here, CD-1 were permitted to interact with C57BL/6J intruders on one side of the CPP setup, whereas the non-conditioned side was empty. AGGs show a positive change in preference under these conditions where NONs showed aversion – they kept away from the intruder. This suggests that AGGs find the aggressive stimulation rewarding and actively seek out the interaction with C57BL/6J.

The authors next examined the neural circuits that control this behaviour with a focus on the connection between the basal forebrain and lateral habenula. During formation of aggression mediated-CPP AGGs show increased forebrain activation (in the nucleus accumbens, diagonal band and lateral septum) with a simultaneous reduction in habenula activity. Next, state of the art optogenetic techniques were used to either activate or inhibit the habenula. Stimulation of the forebrain-habenula circuit in NONs caused a positive change in place preference, whereas inhibition of this circuit in AGGs decreased CPP: the valency of the C57BL/6J stimulus mouse had changed! Importantly, optogenetic stimulation did not alter other social behaviours, although both the intensity of aggression and the rewarding properties of cocaine were increased.

This is exciting research that has possible translational potential to other species. The habenula (and in particular its interaction with the forebrain) appears to be important in processing of stimuli that can elicit aggression. In humans, deep brain stimulation of the basal forebrain and habenula has already been achieved suggesting a possible future treatment for pathological aggression.