How oxytocin regulates social reward

Why do we care about other people? Not just why do we care for them, but why do we care about – their existence? their presence? what they do and how they make us feel?

For a long time, the canonical explanation has been that the hormone oxytocin is a sort of ‘love hormone’ whereby release causes some sort of bonding between two individuals. This story comes to us from the gerbil-like prairie voles. Prairie voles, you see, are pair-bonders who hook up with one mate for life. They’re so attached to each other that once bonded, males will attack any new female that they see (so much for “love” hormone). Luckily for us scientists, there is another closely related vole that does not pair bond. This made it relatively easy to trace the difference: oxytocin receptors in the nucleus accumbens (NAcc).

The NAcc is an area of the brain that is directly involved in motivation and reward; we tend to think of it as the place where the brain keeps track of how rewarding something is. By acting as a sort of central coordination center for value, it can directly promote physical behaviors. Activating the correct neurons related to reward on the left side of the animal will cause the animal to physically turn to the left.

The bond that prairie voles form is linked to oxytocin receptors in NAcc that change neural activity (and I’m simplifying a bit by neglecting the role of the related hormone vasopressin). This change makes their social (pair-bonded) life more rewarding.

At least, that’s one view. But many animals have a social life that does not involve pair bonding, and often they do not have oxytocin receptors in their NAcc. If oxytocin in NAcc was required for strong social behaviors, if they don’t have the receptors how do they have social behaviors at all?

In what I consider the most exciting paper so far this year, Dölen et al investigate what is the neural circuit that makes social interactions rewarding. Mice are actually social creatures, living in small groups to share parental and defensive responsibilities. Dölen et al exploit this by using a variation on a classic conditioned place preference (CPP) experiment. Mice are placed in one identifiable room with other mice (social); they are then placed in another identifiable room on their own (isolated). When they are finally put in a box with two rooms, one that looks like their social condition and one that looks like the isolated one, they spend much more time in the room that reminded them of their social experience. We tend to think this means they prefer that room because it was somehow more rewarding (or less aversive).

This social conditioning requires oxytocin. Yet, when they delete the oxytocin receptors from cells in NAcc animals still become conditioned. It is only when oxytocin receptors in other areas that project into NAcc do they animals lose any social reinforcement. These receptors are in one specific area, the dorsal raphe nucleus, which is a major source of serotonin in the brain. Interestingly, serotonin is also linked to social behaviors and modification of reward circuitry.

What this suggests is that oxytocin affects reward through serotonin; blockade of certain serotonin receptors in NAcc also abolishes social conditioning. It is not surprising that oxytocin could regulate reward in multiple ways. Serotonin may represent distinct aspects of reward – on different timescales, for instance – than other cells that feed into NAcc. By modulating serotonin instead of NAcc itself, oxytocin can precisely fashion the rewarding effects of social behavior.

As a technical matter, they also propose the receptor that serotonin is acting through (5HT1B). I am under the impression that this is an autoreceptor in NAcc. In other words, it is on the serotonin-emitting cell in order to monitor how much has been released to sculpt the output. By using pharmacology to block the receptor, I worry a bit that they are not getting the receptor which oxytocin is acting through per se but just modifying serotonin release in a gross manner. I feel a little vindicated in this worry by the fact that some of their technical results do not appear to be wholly blocked by 5HT1B blockage.


Dölen G, Darvishzadeh A, Huang KW, & Malenka RC (2013). Social reward requires coordinated activity of nucleus accumbens oxytocin and serotonin. Nature, 501 (7466), 179-84 PMID: 24025838

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s