Friday, September 23, 2011

Patience may be a virtue, but it is also really hard.

My dog eats two meals a day, one in the morning and one in the evening.  She is trained to sit until the bowl is on the floor before she leaps to start eating.  She does it, but it is a real challenge.  She gets excited from the moment that she sees the food being put in the bowl, and it requires a tremendous act of will to keep from leaping to the bowl before it is set down.

This daily routine has an important lesson for behavior change.

Animals (like my dog) have a hard time waiting for a good thing.  Even if I set things up so that she would get more food in a few minutes if she could wait, she would have a lot of trouble waiting.  The idea that something now is more valuable than even more later is called delay discounting.  Animals of all types have a high discount rate.  That is, they have a hard time being patient.  They would rather have something now than more of it later.  This is true, even when the delay is just a few seconds or a minute. 

So, what does this have to do with you?

As humans, we pride ourselves on having some patience.  The dog can hardly wait to snap at her meal, but we can wait until the entire table is served before digging in. 

That pride may be misplaced.

A study by Koji Jimura, Joel Myerson, Joseph Hilgard, Todd Braver, and Leonard Green in the December, 2009 issue of Psychonomic Bulletin and Review makes the point that we’re a lot more like animals than we’d like to think.  In their study, they brought thirsty people to the lab and gave them choices between a small amount of a drink of their choice immediately (juice or water) or a larger amount some time in the near future.  The wait times ranged between 5 seconds and a minute.  The people in the study were told that a new choice would be given at a fixed interval, so they would not get more chances to make choices if they always chose the immediate option.

People who selected the immediate drink, were given a chance to drink right away.  Those who selected the larger drink had to sit through the delay period before getting the chance to drink. 

People in this study showed a very high rate of delay discounting, just as animals do.  That is, unless people were offered a lot more juice to wait, they tended to select the immediate drink.

What does this mean for you?

These results make clear that we are not that much different than other animals in our ability to delay our rewards.  If there is a small reward in front of us now, it is hard to give it up for the promise of a bigger reward in the future.  Short-term temptations put a strong pull on us.

So, how do you succeed in changing your behavior? 

If you are trying to lose weight and you have the choice between a slice of pie now or being thin in the future, you are almost always going to take the piece of pie now.  If you are trying to stop smoking, the cigarette being offered to you is going to be much more attractive to you than your future good health. 

You have to remove the immediate option from your environment.  If you want to lose weight, don’t look at the dessert menu, and don’t keep sweets at home.  If you want to stop smoking, don’t stand outside with your friends while they have a cigarette. 

Don’t put yourself in the position of having to protect big rewards in the future from the small ones in your path right now.


  1. Link to the paper: (free fulltext)

    I like the concept and "main point" of your post, but I'm a little skeptical of some of the theoretical consequences you draw from the paper. What do "patience" and "temptation" have to do with an edible reward, if only the quantity of the reward is what's at stake? If I'm hungry, my goal is simply to satiate that hunger, not to obtain as much food as possible. If you were to tell me that I'd get watery bland rice if I didn't wait but properly cooked rice in a minute, that might make a difference.

    Moreover, the paper itself is relying on a very small (15-participant) sample size. They have spread this pool of participants thinly across multiple conditions... literally only 2 or 3 participants per condition. In Experiment 2, the delay time was said to vary "randomly", but I cannot see how you would avoid introducing bias into the results with only 15 samples unless they controlled the selection so that there would still be an approximately equal number of participants in each condition (which would be difficult to do without eliminating the "random" aspect!).

    Anyway, I like your point about eliminating any possibility of "quick" rewards in order to focus on long-term goals. Do you know of any studies which show any correlation between doing so and achieving success?

  2. In many ways, this work is an extension of the kinds of studies that Walter Mischel has been doing with kids for years. He developed the 'marshmallow task' where kids are given the choice between eating one marshmallow that is in front of them or getting two when the experimenter returns to the room some time later. The kids who succeed at this task often find ways to hide it from themselves (often by turning around). Follow up studies with kids in those tasks show that the ones who are best able to delay gratification are generally more successful (in terms of education and later jobs) than those who aren't.