MAY 2006
Behaving Badly May Be Natural at the Top
Deborah Gruenfeld discusses the psychology of power and leadership.
Deborah Gruenfeld, professor of
organizational behavior and Morgan Stanley director of the Center for Leadership
Development and Research
Photo by Debra
McClinton
by Marguerite Rigoglioso
Q: You are a social psychologist who studies the effects of power on powerful people. What can you tell us about this?
I have been studying the psychological consequences of having power for the past seven years, and I have to say it’s definitely the most fun I’ve had doing research. There are just so many good examples of people with power who behave in ways that demand some kind of psychological explanation.
For example, I had a brief career in journalism, and I occasionally met with Jann Wenner, the founder and publisher of Rolling Stone. What didn’t strike me at the time, but now seems remarkable, was what it was like to be in a meeting with him. He had in his office a small refrigerator within arm’s reach of his desk. As far as I could tell, there were only two things in there: a bottle of vodka and a bag of raw onions. While we were meeting, he would reach over, open the door, drink vodka straight out of the bottle, and eat onions. What’s striking about it now is that none of us ever said anything to him about this, and he never even offered to share! He seemed to think it was perfectly appropriate to do this in a meeting. And that is, I think, a classic example of what we think is going on with power, which is what we call “disinhibition.”
Disinhibition involves acting on your own desires in a social context without considering the effects of your actions. It implies a heightened sensitivity to your own internal state and also a reduced sensitivity to other’s interests and experiences. It implies action orientation in pursuit of a goal and also the possibility that you might use others as a means to an end.
For most of us, in situations where we don’t actually have a lot of power, we walk around with all these different internal desires, motives, drives, and we think about whether it’s appropriate to act on them based on the situation. So most of us navigate the social world by trying to regulate our impulses and control our behavior. But when people experience power, they stop trying to control themselves; they forget that there are social consequences. In fact, arguably, there are no social consequences.
We believe the reason power leads to disinhibition is that power activates the behavioral approach system. This is a psycho-physiological system that regulates our behavior in response to rewards and opportunities. It regulates our responses to food, sex, and opportunities for social dominance. When you’re having an approach response, there are certain regions of the brain that are engaged and others that are shut down. You have a general rushing of the blood from the brain into the extremities to prepare the body for action. So what you get with behavioral approach activation is a body that is relatively well prepared for acting but has a relatively low capacity for thinking in certain kinds of ways.
Q: And how do people behave when they are with a more powerful person?
On the low-power side, what we believe happens is activation of the behavioral inhibition system. This is a psycho-physiological system that regulates our response to threats in the environment, and it involves a different set of neurotransmitters and a different brain region. The inhibition kind of response is what I think of as a deer in the headlights response. What’s happening is that your body is preparing you to have a heightened attentiveness to what others are doing and how you’re being evaluated.
There is some evidence supporting this, particularly from research on animal behavior. We know, for example, that there are distinct differences in serotonin levels based on position in the hierarchy, that serotonin levels change as monkeys move into alpha or beta position.
Q: We doubt you measure serotonin levels in powerful people, so how do you conduct your research?
By training I am an experimental social psychologist, which means I conduct experiments on human subjects to test hypotheses. What I do is I take a large group of people, a sample of some kind of population, bring them into the lab, and divide them into two groups. Half the people are randomly assigned to the power condition, which means I do something to create the experience of them being in power. The other half are assigned to a control condition. They get some other kind of treatment that provides a basis for comparison. To the extent you can identify differences in their behavior in this sort of situation, you can draw conclusions with a relatively high degree of confidence that the differences you’re observing are due to the conditions you’ve created. You can attribute those differences to power, as opposed to something else like personality, for example, or some other demographic characteristic. So we design the experiment to try to look at the effect of a particular variable, controlling for the effect of other variables.
We have an interesting study where we look at behavior in a game of blackjack. We brought people into the lab and had them playing blackjack on the computer, and we created a situation where they had 16, which is a kind of weird place where it’s unclear whether it’s good to take another hit or to stay. We found that when they were faced with this kind of ambiguous situation, people in power were more likely to take a hit than people with less power.
In what we called “the E study,” we put participants into conditions of either high or low power and asked them, among other things, to draw a capital E on their foreheads with a wash-off, nontoxic Crayola marker. Think about it. Try to imagine yourself drawing a capital E on your forehead. If you want the E to be legible to yourself, you draw it one way, backward to the audience. If you’re concerned about whether your audience can read it, you draw it the other way, so it’s legible to your audience but backward to you. The idea was that if power is reducing concern with how the audience might perceive you, then people in our power condition will be less likely to spontaneously think of the possibility that someone else would have to read the E, and that’s exactly what we found.
One of my favorite studies, conducted by one of my co-authors, used groups of three undergraduates who were told they were participating in a study of their attitudes about a number of different social issues. They were to discuss each issue, come to a group position, and write a brief essay on each one. Before they started, one of the three was chosen at random to evaluate the performance of the other two and assign points that reflected how well they performed. The points mattered because they went into a lottery that affected their chances of winning a cash bonus when the experiment was over. About 30 minutes into the experiment, the experimenter came into the room with a plate of cookies. There were five cookies on the plate and three people in the room. Everyone can take one cookie with no tension, there’s nothing awkward there, but then you have two cookies left and three people in the room. What we found is that not only did the participants in high power eat more cookies, but they ate with greater abandon and were more likely to get food on their faces and crumbs on the table!
If you have not had the pleasure of dining with a disinhibited eater yourself, I want to share this example that I think is kind of fun. It’s from Bob Woodward’s book Plan of Attack.
“The Joint Chief’s staff had placed a peppermint at each place. [George W.] Bush unwrapped his and popped his into his mouth. Later, he eyed Cohen’s mint and flashed a pantomime query, ‘Do you want that?’ Cohen signaled ‘no,’ so Bush reached over and took it. Near the end of the hour-and-a-quarter briefing, the chairman of the Joint Chiefs, Army General Henry ‘Hugh’ Shelton, noticed Bush eying his mint, so he passed it over. Cheney listened, but he was tired and closed his eyes, conspicuously nodding off several times. Rumsfeld, who was sitting at a far end of the table, paid close attention, though he kept asking the briefers to please speak up, or please speak louder. ‘We’re off to a great start,’ one of the chiefs commented privately to a colleague after the session. ‘The vice president fell asleep and the secretary of defense can’t hear.’”
The basic idea we’ve tried to show, what we believe is going on, is that power is disinhibiting and it activates the behavioral approach system. As a consequence, powerful people are more action-oriented even when it’s unclear whether acting is permissible or socially appropriate. They’re less concerned with what others think of them and their behavior and less cognizant of what others think of them in general. As a result, they’re more inclined to see the social world through the lens of self-interest and to view people as instruments for the accomplishment of personal goals.
That said, I want to be clear that I’m actually agnostic as to whether power is a good or bad thing. Disinhibition in response to power has positive consequences as well as negative. Social and psychological research shows that whenever you see these kinds of tendencies associated with certain positions in the social structure, they probably have some kind of adaptive function. So I want to acknowledge that there are, I’m sure, some very functional reasons that people with power tend to behave in these ways.
When people experience power, they stop trying to control themselves; they forget that there are social consequences.
