The rewarding nature of social contact

See allHide authors and affiliations

Science  29 Sep 2017:
Vol. 357, Issue 6358, pp. 1353-1354
DOI: 10.1126/science.aao7192

How is the pleasure of social interaction wired in the brain?


Some of the most fascinating questions in theology and philosophy are now being tackled by the social and affective neurosciences. For example, are people inherently good? Are our capacities for love, empathy, and altruism uniquely human? Are we designed for monogamy? On page 1406 of this issue, Hung et al. (1) provide an important step in a long history of scientific attempts to address these questions about our social nature.

The fundamental components of most human traits are shared by other species, in both form and function, indicating a common ancestor and an evolution over time (2). For example, humans, monkeys, rodents, sheep, and dogs all share mechanisms for bonding with mates or for protecting newborn kin—processes that support acts of kindness (3). Indeed, the famous social bonding neuropeptide oxytocin can be traced over 500 million years, with analogous peptides found in birds, reptiles, fish, amphibians, and some invertebrates (4).

Converging research on the role of oxytocin in social bonding suggests that approaching others becomes less scary and more rewarding when it is valuable to the individual. In the brain, oxytocin is secreted by the paraventricular nucleus (PVN) of the hypothalamus and projects to regions that are rich with oxytocin receptors, including those associated with dopaminergic reward-based motivation [e.g., ventral tegmental area (VTA) and nucleus accumbens (NAcc) in the mesolimbocortical system]. For example, the density of oxytocin receptors in the NAcc is higher in monogamous than nonmonogamous voles, and the monogamous bond is prevented if oxytocin receptors are blocked or NAcc dopamine is antagonized (5). Oxytocin also facilitates the motivation to approach newborn pups through connections to the dopaminergic VTA (6). For bonded mates and offspring, oxytocin facilitates approach while helping to establish the memory of the partner and reinforcing the bond. Because the dopaminergic reward system is also linked to motivation for hedonic pleasures and reinforcement (including drugs, sexual experiences, chocolate, goods, and attractive faces) (7), it has been assumed that oxytocin facilitates social bonds by rendering another individual like a drug—something to approach, enjoy, remember, and seek again.

This interpretation is so intuitive that hundreds of “popular science” articles have been written about oxytocin as the “love drug” or the “hug/trust hormone.” However, the reality is far from being this simplistic. For example, there are multiple anatomical systems of dopamine projections and multiple receptors for dopamine that have distinct properties. Moreover, oxytocin distributions and mechanisms change across highly related species, between males and females, throughout the life span, and even according to social context.

The data in Hung et al. point to both the simple and the complex features of the oxytocin framework, as they examined how dopaminergic neurons in the VTA support social preferences. They used a diagnostic behavioral test to determine the degree to which male mice preferred to spend time in a chamber with an unknown juvenile male mouse inside (the social preference). They identified oxytocin neurons that project from the PVN to the VTA; inhibiting these neurons or knocking out the VTA oxytocin receptors reduced the social preference. These interventions did not affect the ability to move about or to prefer a chamber associated with nonsocial reward (a cocaine chamber), meaning that the oxytocin-VTA mechanism may only facilitate social forms of reward. These VTA-projecting PVN oxytocin neurons increased activity when mice interacted with the unknown juvenile but not with a toy mouse. Photostimulating PVN oxytocin neurons could not itself reinforce a preference for an empty chamber or reinforce a nonsocial nose-poke behavior; this indicates that the oxytocin release itself is not pleasurable. The authors found increased spontaneous cell firing in NAcc-projecting VTA dopamine neurons after applying an oxytocin receptor agonist. Hung et al. concluded that oxytocin release in the VTA from the PVN is linked, at the time of a new social interaction, to increased excitability of dopamine neurons that project to the NAcc, releasing dopamine in the NAcc and, thereby, reinforcing the social interaction.

Because this topic tends toward overstatement, a few caveats are warranted. The Hung et al. studies were performed in male lab mice, which are not monogamous. Thus, the social reward came in the form of preferring to meet an unknown juvenile male, rather than a familiar or bonded partner, as is more typical in rodent social bonding studies. The mechanisms for oxytocin and reward are known to differ between monogamous and nonmonogamous species, between caregivers and noncaregivers, and between males and females (6). Thus, the exact neural dynamics may shift in another species or context. However, that Hung et al. found an oxytocin-VTA link in a promiscuous male mouse perhaps indicates a conservative estimate of the importance of this link in social interaction.

Some of the terms in this paper must be interpreted carefully. Hung et al. use “reward” to refer to dopaminergic NAcc mechanisms, which support motivating and reinforcing behaviors. However, oxytocin release was not associated with feeling pleasure per se, and dopaminergic processes even within the NAcc shell can represent multiple feeling states (8). Thus, reward should be interpreted as a motivation to meet an unfamiliar animal, which may also be remembered and reinforced.

The term “prosocial” is used here to mean “for social behavior,” whereas in humans it is only used when individuals are particularly social or generous (e.g., in altruistic giving). These mice did not help one another. However, such aid has been demonstrated and linked to oxytocin in rodents and humans (2). Oxytocin in humans has also been linked to rewarding and pleasurable phenomena such as romantic love, parenting, and comforting touch; and altruism may be promoted by this oxytocin-VTA mechanism (3). However, there are many noted failures in human research to replicate associations between oxytocin and prosocial behavior (911). Such failures may reflect that rodent research usually involves clearly bonded, adaptive contexts (mating and caregiving), whereas human research employs more abstract tasks such as giving money to a stranger.

How social processes become rewarding

Studies in mice suggest that social behavior in humans occurs because of the connections between oxytocin and the reward-based dopaminergic system, which presumably mediates the ability of humans to notice, seek, remember, and return to rewarding experiences of all types—in this case social contact.


Given these complexities, it is refreshing that Hung et al. were able to establish a link between oxytocin and reward in a less clearly “loving” context, but considerable work is required to reveal the “devil in the details” in this popular, and yet unpredictable system.


View Abstract

Stay Connected to Science

Navigate This Article