No love for outsiders – oxytocin boosts favouritism towards our own ethnic or cultural group

By Ed Yong | January 11, 2011 9:00 am

Few molecules have a reputation as glowing as that of oxytocin. Often billed as the “love hormone” or “cuddle hormone”, oxytocin has been linked to virtually every positive aspect of human behaviour, including trust, social skills, empathy, generosity, cooperation, and even orgasm. And to this extensive list, we can now add racial and cultural bias.

Despite its misleading labels, oxytocin has a dark side. Just two months ago, Jennifer Bartz showed that it can make people remember their mothers as less caring and more distant if they themselves are anxious about social relationships. Carolyn H. Declerck found that oxytocin makes people more cooperative in a social game, if they had met their partner beforehand. If they played with an anonymous partner who they knew nothing about, oxytocin actually made them less cooperative. “Oxytocin does not unconditionally support trust,” she says.

Now, Carsten de Dreu from the University of Amsterdam has found that sniffs of oxytocin make us more biased towards peers from our own ethnic or cultural group, versus those from other groups. Bartz commends the new study, saying, “Along with other recent reports, [the new study] suggests that although oxytocin clearly plays a role in prosociality and empathy, the way it does this is more nuanced than previously thought. This is not entirely surprising given the complexity of human relations.”


De Dreu asked 280 Dutch men to take three puffs form an oxytocin nose-spray, or a placebo that contained the same mixture without the hormone. It was a “double-blind” study – neither de Dreu nor the men knew who had been given what until the results were in.

First, de Dreu looked for any hidden biases in the volunteers’ reactions to German, Arab or other Dutch men. He used an ‘implicit association test, where volunteers used two keys to categorise words into different groups (e.g. Dutch names or German/Arab names, or positive and negative). Combinations of categories that contradict our biases should subtly slow our reaction times. If people are biased against Arab people, they’d take longer to finish the test if the same key was assigned to both Arab names and positive words. These “implicit associations” are very hard to fake, especially if the test is done at speed.

Sure enough, oxytocin strengthened the biases of the Dutch volunteers. When they sniffed oxytocin (rather than the placebo), they were quicker to associate positive words with Dutch names than with either German or Arab ones.

Such biases can affect how we see other people. We humanise those who are part of the same group, ascribing a more complex range of emotions to them. By contrast, we have a tendency to dehumanise outsiders, by assuming that their emotional lives are narrower. This particularly applies to so-called “secondary emotions”, such as admiration, hope or embarrassment, which are seen as unique to humans (in contrast to “primary emotions” like happiness, fear or disgust that are common to other animals).

De Dreu found that oxytocin strengthens these tendencies. He asked 66 white Dutch men to sniff either oxytocin or placebo before showing them pictures of other Dutch or Middle Eastern people. The volunteers had to say how strongly the people in the images would experience different emotions. Both groups were more likely to ascribe secondary emotions to people within their group than those outside it, but that difference was even greater after a sniff of oxytocin.

Finally, de Dreu showed that these shifting biases could affect the moral choices we make. He presented volunteers with a famous series of moral dilemmas. For example, a runaway rail trolley is hurtling towards five people who are about to be killed unless you flip a switch that diverts the trolley into the path of just one person. All of the dilemmas took the same form – you weigh the lives of one person against a group. And in all the cases, the lone person had either a Dutch, German or Arab name, while the group were nameless.

After a sniff of placebo, the Dutch volunteers were just as likely to sacrifice the single person, no matter what name they had. But after sniffing oxytocin, they were far less likely to sacrifice the Dutch loners than the German and Arab ones.

This last experiment clearly shows a trend that applied to the whole study: oxytocin boosted favouritism for people who belong to the same group. Only very rarely did it increase negative feelings towards people outside it. For example, in the moral dilemmas, oxytocin made the volunteers less likely to sacrifice members of their own group, but not more likely to sacrifice outsiders.


The stereotypical view of oxytocin is that it increases positive feelings rather than negative ones. But it doesn’t do so equally. Rather than being an all-purpose chemical of social affection, you could view oxytocin as a drug that strengthens our tendency to discriminate between people within and outside our social cliques. “Different” can come under many guises – in de Dreu’s experiments, Germans and Arab people were both treated as outsiders,  despite having very different characteristics and stereotypes.

Bartz notes that oxytocin was simply amplifying the biases that volunteers already had. They already had a degree of favouritism towards their peers and oxytocin enhanced that. To Bartz, this suggests that the hormone could be heightening the importance of social cues in a person’s environment, something altogether subtler than simply making everyone cuddly.

De Dreu’s results fit well with previous studies, including animals ones. Under the influence of oxytocin, rat mothers become more aggressive towards intruders. Genetically engineered mice that can’t respond to oxytocin develop a sort of “social amnesia”, where they can’t tell the difference between mice from their own group, and outsiders. Last year, De Dreu showed that oxytocin drives a “tend and defend” response. where people are more trusting and cooperative towards others within their group, but become more aggressively defensive to people outside it.

This sort of favouritism makes a degree of evolutionary sense. It could bolster trust and cooperation within a community, such that groups whose members stuck together more would out-compete those that did not. If that’s the case, you would expect the brain to have some way of sustaining racial biases and oxytocin could help with that.

But such preferential treatment has an obvious dark side – it leads to all sorts of moral and cultural problems, including inequality, discrimination, prejudice, and conflicts between different groups. As de Dreu writes, “Oxytocin may trigger a chain reaction toward intense between-group conflict. This possibility questions the rather widespread view of oxytocin as a ‘cuddle chemical’ or ‘love drug’.” If it’s a “love hormone”, then that love is a narrow one.

Reference: PNAS
If the citation link isn’t working, read why here

More on oxytocin:


Comments (10)

  1. Nice work, and amid the apparent contradictions I think this points to a key area of consistency: the actual end-product of many of these hormones (and related polymorphisms that sometimes appear to influence their levels of activity) oft depends on context.

    Another worth noting: Culture has some power to define in-groups and outgroups, allowing us to move beyond obvious (apparent) in-group/outgroup markers such as appearance (skin color etc) and focus on shared values or behaviors or practices, such as language etc. This is why, to take a simple example, two fans who favor the sports team might identify each other as part of the same in-group even though they are of different ethnic/racial heritage. Similarly, a recent study, if I remember correctly, found that children paid more attention to similarity of speech than they did differences in race. Culture can and does frame and shape how we actually express these evolutionarily developed traits and tendencies, such as sensitivity to in-group/out-group differences.

  2. Indeed, and check out this earlier post on oxytocin, showing an interaction between gene, environment AND culture.

  3. There’s also the oxytocin as autism treatment study (Andari et al. 2010), which got a lot of publicity. Oxytocin resulted in autistics displaying us-vs-them behaviour, widely applauded as better social skills. I disagreed with this shortsighted and not very prosocial interpretation.

  4. becca

    So what chemical makes me want to cuddle all the little ducks, *especially* the yellow one?!

    This is a pretty cool story. Even though I had a pretty nuanced view of oxytocin (and other hormones), I would not have predicted this outcome. Still, it makes sense in retrospect. Which says something about the importance of the work.

  5. this is the proximate neuro entailment from multi-level selection models where group dynamics matter. in any case, i would add one minor point: evolutionarily this isn’t about racism at all, because interaction of very different races is a modern feature of a global world. rather, these tendencies would be emerging in the context of nearby ethnicities who were probably very similar in most conventional metrics. a biosocial aspect may have emerged because of the necessity for a “push” in that direction. though cultural markers such as accent or tattoos can separate similar groups easilytoo.

  6. Matt B.

    So the distant-mother thing would be explained by saying that she is perceived as an outsider by the child.

  7. But it might not be the oxytocin that is causing the effects independently. Perhaps both sets of qualities (“good” and “bad”) are just highly correlated in people. I only have a sample of one though, for my ridiculous theory. As a (thinking rather than feeling) unempathetic woman with naturally bad social skills, I have noticed that my empathetic sisters (ie other women) tend naturally towards inequality, discrimination and prejudice by showing amazing generosity to their in crowd, at the expense of others. And they are so “intuitive” that they don’t even realise it.

    Well – just my experience, but I think we need to remember that society needs the unempathetic thinkers as well as the touchy feely huggers, so probably best we stay off the oxytocin pills!

  8. Its interesting how we forget that most of us when presented with novel situations have natural affinities to those who look more like us. I would suspect that this is due as much to our biological drives for relationship as it is specifically a negative expression of oxytocin. Oxytocin is the anti-stress hormone thereby it modulates amygdala activity during stressful encounters but this does not indicate that it creates or triggers any form of cultural bias that may not have already been unconsciously present. There are numerous studies available at Oxytocin Central.

  9. “WE” know so little of the human systems that relate to friend or foe—-fear & pain–fight or flight and it is probably 100 years to early to be establishing any patterns be they social or genetic—not enough time & resources are , have been or will be spent on self enlightenment –we are doomed to self destruction , between greed and fighting be it nobility or clergy the commoner will always lose and humanity’s very existance is in constant peril. A point to ponder is the fear in slavic people have— a consistant fear of other people because of the terror suffered from centuries of mongul extermination raids—is embedded in their brains .

  10. these are studies done with of dutch men. where are the studies with women who have more naturally occurring oxytocin? the results and conclusion drawn are ridiculous..of course bonding within group will be generated as needful for survival. unempathic evaluation is unbalancced.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar