Why the “us and them” mentality evolved

Humans have been living in large complex groups for millions of years, and flourishing as a result. This social environment has shaped the course of our evolution as much as any natural one, in particular how our psychology has evolved. For example, humans are highly susceptible to peer pressure, likely to ensure we do what the group wants and so don’t risk being thrown out of it. A human going solo would not survive for long (Hawkes et al., 1991).

Now new research has identified yet another aspect of human psychology that seems to have evolved to help us deal with living in social groups: dividing the world into an in-group and an out-group, and working extra hard to curry favour with the former. Although the habit of humans to break the world down into “us” and “them” has been known about for years (and often blamed for many sources of tension in the world) there has always been debate over whether it is an innate psychological tendency or something we are trained to do (van Vugt and Hardy, 2010).

After all, we’re exposed to loads of things that train us to split the world into in- and out-groups, from sports teams to countries to religions. But are they the source of this division, or simply the by-product of us naturally breaking the world down. This new research believes they’ve solved this mystery by investigating whether very young children (5 years old) also divide up the world like this.

First they gathered up a lot of children from various different day care centres so they wouldn’t know each other. They then dressed them up in either red or blue and put two of them in a room. One was given a set of stickers and asked to donate some to an unknown third child who would enter the room later, the other told to observe. If the observer was wearing the same colour as the donater then they were much more generous, giving twice as many stickers to the unknown child (Engleman et al., 2013).

The in-group and out-group scenario these kids were presented with

The in-group and out-group scenario these kids were presented with

This is a similar to how adults behave in a similar situation, being more generous when watched by in-group members in an attempt to build a reputation with them (Reiss and Gruzen, 1976). However, since this study was dealing with children who have yet to fully mature psychologically (Wimmer and Perner, 1983) and haven’t been exposed to quite as many institutions trying to make us split the world up into “us and them.” As such this strongly indicates that they will not have been influenced by culture in quite the same way, suggesting that this behaviour is at least partly biological and thus evolved.

Ordinarily speculating about the influence of evolution would be unjustified, it’s where evolutionary psychology often goes very wrong. However, this study identified an additional interesting effect that may shed light on the evolution of the “us and them” mentality.

The researchers repeated the experiment, but sometimes the children involved were told that the observer would then become the donater, and the donater would be the one relying on his charity to get stickers. When this was done the original donater would behave generously, regardless of whether the observer was wearing the right colour or not (Engelman et al., 2013).

This suggests that they key motivation behind dividing the world up into in- and out-groups was reciprocity. You acted nicely towards members of an in-group because they were part of your group and thus you were more likely to encounter them later. As such, if you build up your reputation with them by being generous, they may be more likely to help you when you see them around town. So if you know a member of the out-group will be in a position where they could help you later, it pays to curate a reputation with them too.

Unfortunately this research alone is not enough to put this issue to bed. Speculation is not EvoPsych’s only problem, it also tends to extrapolate results from a small group of humans, claiming that the phenomena identified in them are present in all humans and thus innate and evolved. Whilst this research did involve children from several daycare centres, it’s still not enough to justify such extrapolation.

If these results are confirmed by further studies then this research could help us understand one of the biggest sources of tension in the world. Which will hopefully lead to a better future, as well as giving us a greater understanding of our past.


Engelmann, J. M., Over, H., Herrmann, E., & Tomasello, M. (2013). Young children care more about their reputation with ingroup members and potential reciprocators. Developmental Science.

Hawkes, K., O’connell, J. F., Jones, N. B., Oftedal, O. T., & Blumenschine, R. J. (1991). Hunting income patterns among the Hadza: big game, common goods, foraging goals and the evolution of the human diet [and discussion].Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences334(1270), 243-251.

Reis, H.T., & Gruzen, J. (1976). On mediating equity, equality, and self-interest: the role of self-presentation in social exchange.Journal of Experimental Social Psychology, 12, 487–503.

van Vugt, M., & Hardy, C.L. (2010). Cooperation for reputation: wasteful contributions as costly signals in public goods. Group Processes & Intergroup Relations, 13, 101–111.

Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception. Cognition13(1), 103-128.

16 thoughts on “Why the “us and them” mentality evolved

  1. Pingback: Arkarbor

  2. I enjoyed reading your post. Robert Sapolsky, in his article “Peace Among Primates” wrote:

    Humans may be hard-wired to get edgy around the Other, but our views on who falls into that category are decidedly malleable (neuroplasticity).”

    Dr. Sapolsky’s comment was based on observation after spending years among savanna baboons, and experiments using fMRI scans on humans. When the subjects were shown faces of someone from a difference race (for example), the amygdala got metabolically active (alert, aroused and ready for action). This also happened when faces were presented subliminally. However, the right amygdala didn’t activate with subjects who had a lot of exposure to different races. In other experiments, when they subtly bias the subjects beforehand to think of people as individuals rather than members of a group, the amygdala didn’t budge, either. There is hope. =)

    • So it would seem the trick to ending prejudice is showing how helping people can benefit oneself, and pointing out that those other people are still people.

      It seems painfully obvious, it worries me slightly we haven’t managed to do it yet

    • This stance has been most prominently promoted by Peter Singer as the “expanding circle”: if we can’t overcome our ethnocentric predisposition then why not expand the in-group to include all of humanity? It seems so painfully simple and obvious. Unfortunately, solutions to long standing social ills are seldom so simple. Even the experiment discussed by Adam makes this painfully clear: the kids weren’t taught in daycare or by society to discriminate based on the colour of people’s hats; yet when given such signals, they used the most restrictive and salient signal they could to bias their decision making. To achieve an egalitarian society we need to think harder than just blind hope that people’s better natures will prevail and they will naturally expand their circles.

  3. The experiement is full of assumptions, and I don’t buy the main ones. That kids of that age have not been exposed to cultural influences the same way an adult has is rubbish, and I can’t see why they assume the result can be extrapolated to all of humanity as a built-in evolutionary future – that is just fantasy. Also, it is obvious to almost anyone who has ever been a kid that kids think in ingroup/outgroup categories. The experiment just confirms common sense and then extrapolate it to imply a new discovery with ground breaking implications.

    The one part of it I find interesting is the reciprocity aspect in the kids – that expectation of a future reward was just as motivating as making a good impression on an ingroup members (but again, it is just one of their assumptions that that was the motivation behind the behaviour).

    Building on that assumption, it could be interesting to set up an experiement with a conflict between those two parameters – expect of future reward from outgroup member VS making a good impression on ingroup observer – and see what the kids chose.

    • I don’t think the researchers were trying to claim that the children were devoid of cultural influence, but that it would be significantly different to that adults are regularly exposed to. Thus the fact there is a similar effect documented in both circumstances suggests culture alone cannot explain it.

      Also, whilst research into children has identified this effect before it was only documented in those as young as 8. Whether the fact they’ve pushed it back to age 5 is significant I’ll leave for the child psychologists to decide.

      Beyond that, I agree with pretty much everything you say. There’s interesting potential here, but more work is needed

  4. Pingback: » Why the “Us and Them” Mentality Evolved | EvoAnth Let's Get Political

    • To draw attention to the fact they were all part of the same group, the researchers got the children to identify who was wearing what coloured hat.

      So the answer is clearly blue hats, duh!

  5. “If the observer was wearing the same colour as the donater then they were much more generous, giving five times as many stickers to the unknown child ”

    I think that this is incorrect. If I understand which result you are referring to, you’re looking at the difference between the outgroup and ingroup treatment conditional on the no reciprocity condition, right?

    In the paper, they seem to say that the donaters in the outgroup treatment give 1.6 stickers. Those in the ingroup give 3.5 stickers, so they donate only twice as much. The difference still appears significant though.

  6. I think a lot of the research in this field conflates two different types of us-them effects. They tend to confuse in-group favoritism and out-group hostility. It might seem like the two are somehow equivalent, but I think a distinction needs to be made, especially in the context of evolutionary game theory, for instance. This study, for example, shows in-group favoritism: i.e. in the one-shot treatment, the ‘best thing’ to do for the individual is to not give any stickers and keep all for yourself. This is often easier to show that out-group hostility, which is the real culprit of bad things in the world; that being said I published a study in 2010 showing that evolution can lead to either in a game-theoretic context. With in-group favoritism (the usual thing studied by game-theoretic methods), the judgement of ‘better’ between an egalitarian and non-egalitarian society can get a little bit more values based. It starts to depend on if you are a humanist or a utilitarian. Obviously, in the out-group hostility setting, both the humanist and utilitarian would agree that out-group hostility is bad (although I have yet to show this in simulation).

    The relationship to reciprocity is completely tangential in this study, and I think the conclusion they present is unwarranted. Just because people act rationally in a reciprocity setting (i.e. cooperate so other will be more likely to cooperate back with you), doesn’t explain anything about the more irrational form of cooperation seen in one-shot encounters (Although again, the question of ‘rationality’ in such settings can be much more subtle). I think it is also silly to suggest that individuals actually think about reputation in the group explicitly in such settings, as opposed to unconscious biases or even quasi-magical thinking.

    • You hit on a good point there. Their original experiment was a one shot, with a single interaction occurring between the participants. Their second experiment pointed out to the children that there would be multiple rounds (albeit with roles being switched) and they don’t seem to have controlled for this shift in game design. This raises questions over whether it was the change in the nature of the game or the fact roles would be changing that caused the observed effects in the second experiment. Particularly given, iirc, being more co-operative becomes more beneficial in a multi round game anyway, regardless of whether or not other participants have a chance to observe you.

  7. Pingback: V for Versus | Nothing but hoopla

You evolved too. Have a say.

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

WordPress.com Logo

You are commenting using your WordPress.com 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