Why hierarchy matters

During the dot-com boom, it was common to hear about companies with “flat hierarchies” to more flexibly meet new challenges. No longer was there one long chain of command: now most everyone was equal to everyone else. I don’t know whether that concept still exists, but my recent reading into the science of social status has led me to reconsider my previously full-throated support of the concept. Here are some basic reasons why hierarchies are good:

Human beings are social animals, a fact that is central to how we as a species see the world. And like other social animals, whether wolves or chickens or chimpanzees, we sort ourselves into rankings. These rankings aren’t static, they can change over time, but they impose order on social interaction: In the wild, they create a framework for dividing up vital tasks among a group, and because they clearly codify differences in power or strength or ability, they prevent every interaction from disintegrating into an outright fight over mates or resources — someone’s rank tells you how likely she is to beat you in a fight, and you’re less likely to bother her if you already know.

…For one thing, it turns out that people are ruthlessly clear-eyed judges of their own place in the social hierarchy. This is notable because they tend to be poor judges of just about everything else about themselves. Study after study has shown that people are incorrigible self-inflaters, wildly overestimating their own intelligence, sexual attractiveness, driving skills, income rank, and the like. But not social status, that they turn out to be coldly impartial about.

For example, a team of social psychologists led by Cameron Anderson of the University of California, Berkeley ran a study in which strangers were put into groups that met once a week, and were tasked with solving various collaborative problems. After each meeting, the participants rated their own status in the group and that of their teammates. By and large, people’s self-evaluations matched up with how their peers rated them.

…In a 2003 study by Larissa Tiedens and Alison Fragale, both then at Stanford University, subjects who displayed submissive body language were found to feel more comfortable around others who displayed dominant body language than around those who also displayed submissive body language — and to like those with more dominant posture better, as well. People, it seems, prefer having their evaluation of social hierarchy confirmed, even when they see themselves at the bottom of it.

Perhaps the strongest, if the most surprising, evidence for the importance of clearly delineated social hierarchies is work that suggests that more inequality can make for better teams…Galinsky and his fellow researchers found that NBA teams with greater pay disparities not only won more, but ranked higher in categories like assists and rebounding, suggesting a higher degree of cooperation. The clearer the status imbalance, the researchers argued, the less question there is about where one stands.

Now clearly inequality per se is not a good thing. But there are situations where inequality between individuals is good. We are intensely social animals and function better as a group when we have a status; it’s better to have a low-entropy social pattern than a high-entropy one. Social status even affects us biologically; it changes the number of dopamine receptors we have in the basal ganglia which (should) affect how we value things and how pleasurable activities are. Inequality, it seems, is in our very brains.

Social status, novelty-seeking, and dopamine

Dopamine sure seems to do a lot of things these days, doesn’t it? It’s most commonly thought to be the mechanism for prediction error which is used for reward-based learning, but it is also linked to sociability, pain, and a ton of other things.

One of its mechanisms relates to social dominance. Eight years ago, Morgan et al. published a paper concerning social dominance in monkeys. They wanted to show how there is a profound environmental influence on dopamine function, so they separated monkeys into housing blocks of four apiece. Whereas before there was little difference in dopamine (D2) levels between monkeys, after three months of living together the monkeys that had become dominant showed a more than 20% increase in dopamine receptor density, with the most submissive showing no change. Unrelated to our point, but still interesting, they also showed that the submissive monkeys were much more inclined to self-administer cocaine; the dominant monkeys were much less likely to do so.

Dopamine receptor density is therefore an environmental and cultural phenomenon – at least in monkeys. What about people? It turns out we’re pretty much the same way. If you do the same PET scan on people that you did on monkeys, their dopamine level (D2/3 in the striatum) correlates (r^2 ~ 0.5) with a measure of human social status. So we’re not that different, after all. Just remember it’s not dopamine that drives dominance, but dominance that drives dopamine. Why? Who knows?

Dopamine levels in the same area are related to other aspects of behavior besides just dominance. If you perform the same measure of dopamine receptor availability but compare it to sensation-seeking, or novelty seeking, you get an altogether different type of curve. Here we instead see an inverted U-shaped curve. Using modeling, they suggest that those with the greatest need to seek sensational activities have both low receptor availability and high dopamine occupancy. I suppose the low-sensation seekers have low receptor availability and low dopamine occupancy (ie, few binding sites along with a lower volume of dopamine to bind).

Are these two phenomena related? They both look at the receptor availability of a certain population of dopamine receptors (D2/3). But just because the dopamine receptors are the same and in roughly the same place doesn’t mean the connectivity is the same. Are these all the same circuit? Does dominance affect novelty-seeking? Someone really needs to find an animal model to start testing these propositions.

References
Gjedde, et al., 2010. Inverted-U-shaped correlation between dopamine receptor availability in striatum and sensation seeking. Link.

Martinez, et al. 2010. Dopamine type 2/3 receptor availability in the striatum and social status in human volunteers. DOI

Morgan, et al. 2002. Social dominance in monkeys: dopamine D2 receptors and cocaine self-administration. Link.

[monkey photo from]