The Psychology of Irrationality

Notes and comments on a chapter by Roy Baumeister, a psychologist at Florida State who studies self-control and decision-making.

He identifies three basic categories of irrational and self-defeating behavior:

(1) Deliberate self-harm, for which he says he has found no evidence in normal adults.
(2) Trade-offs: when something good is linked to something bad.
(3) Counterproductive strategies: a way of trying to get something good that doesn't work.

The primary causes of self-defeating choices:

(1) Emotional distress: Lottery experiments* suggest that people who are upset tend to make foolish, risky choices because they fail to think through the implications and consequences of their behavior. In his research, when people who were manipulated by the experiment to become angry are told to take a minute to "stop and think" before making a choice, they make better decisions, even though they do not calm down in such a short time. Angry people took riskier choices when that gave them either a lower or higher expected gain.

*Lottery experiments are very common. There are a lot of variations, but generally in these experiments, people are given the choice between two or more options that have different amounts of pay-off (e.g. $5 or $50) and/or different probabilities of winning the amount (e.g. 50% or 5%). A "rational" person will generally choose the option with the highest expected value, calculated by multiplying the probability by the pay-off amount, subject to a certain level of risk aversion. If the values are the same, they will choose the one with the highest probability. An example of a bad risky choice would be to select a 10% chance at $10 (expected value $1) instead of a 50% chance at $5 (expected value $2.50, plus it's more certain.)

(2) Threatened egotism: "People who hold a high opinion of themselves often get quite upset by a blow to pride and the rush to prove something great about themselves overrides their normal, rational way of dealing with life." He did some interesting experiments involving giving people a fake creativity test, telling them they either scored well or scored poorly, and then giving them the option to bet on their own performance in a task. When told they had done well on the test, high self-esteem people made more money than low self-esteem people - they correctly gauged how well they were going to perform and bet accordingly. But when they were told they had done poorly on the test, they bet more and lost more than the low self-esteem people - they bet more extravagantly even when they did not have the ability to back it up.

(3) Self-regulation failure: Self-regulation refers to how people manage and control themselves. Self-regulation is a limited resource; if it gets depleted, the self cannot function as effectively for some time after that. In his research, people showed up at the lab after having been told to skip lunch and were put in a room with cookies; some people were told to not eat the cookies and were later given radishes instead while others could eat the cookies. The ones who avoided the cookies gave up faster when asked to do either some impossible geometric figure tracings or a set of solvable anagrams than the cookie-eaters did.

In recent research, he has found that self-regulation depends on glucose as an energy source. No wonder dieting is so fricking hard: making the effort to control your food intake makes your glucose level drop, so you want more food.

(4) Decision fatigue: Making decisions, either one large one or a series of small ones, draws on the same resource as self-regulation.

(5) Rejection & belongingness: He believes that the need to belong and to form and maintain connections to others is a more powerful motive than it is often credited with, more powerful than self-esteem, for example. In an experiment, he manipulated someone's feeling of connection by telling them that they were either chosen to be a partner by everyone in a group or chosen by no one (and in another variation, that on a fake test they took, they would be likely to be well connected, or alone in life, or accident-prone: another "bad" outcome that doesn't relate to belongingness). The ones that were made to feel rejected or alone performed less well on many different critieria - IQ test performance, aggressiveness, willingness to help someone who asks, and a variety of "taking care of oneself" measures; they were more likely to eat fattening food, take a risky lottery, and drink less of a vinegary yet "healthy" drink in experiments.

One of the big takeaways of this research to me is that it's important to not set yourself up for failure by lining up too many decisions or too much of a need for self-control at one time. It's basically common sense that it doesn't make sense to do too many hard new things at once: going a diet, start exercising, try to meet a new bedtime hour, give up sugar-free soda for water, cut back on TV viewing, etc. But if self-regulating and decision-making capacity is a limited resource, you need to be sure to protect yourself from self-control lapses when you have big decisions you are contemplating or a bunch of small decisions that have to be made. If you've had a rough day in that respect, you are unlikely to be able to go to an Italian restaurant for dinner and order the healthy tuna dish that you always plan to get; if the lasagna always calls your name, you will be especially vulnerable. Perhaps it's better to go someplace that you don't have to control yourself so much in. And it's no wonder that many people have trouble with sugary snacks at work. Though doesn't it seem like women indulge in this more than men do? I wonder how the glucose theory lines up with other research (e.g. in Brian Wansink's lab) that has found male/female disparities in snacking.

References:

The psychology of economic decisions / edited by Isabelle Brocas and Juan D. Carrillo.
New York : Oxford University Press, 2003

Gailliot, M.T., Baumeister, R.F., DeWall, C.N., Maner, J.K., Plant, E.A., Tice, D.M., Brewer, L.E., & Schmeichel, B.J. (2007). Self-control relies on glucose as a limited energy source: Willpower is more than a metaphor. Journal of Personality and Social Psychology, 92, 325-336.