The XP France list has been talking about this idea the past week, but it's due to the language barrier I'm going to paste a snippet about Kohn's book instead (it came from James Shore's blog http://jamesshore.com/Blog/The-Stunning-Truth.html
a couple years ago):
"as reported by Alfie Kohn in his book Punished by Rewards.
In 1961, a graduate student at the University of Kentucky found something she didn't expect. ...She brought 72 nine-year-old boys into her laboratory one at a time and challenged them to tell [two] faces apart. Some of the boys were paid when they succeeded; others were simply told each time whether or not they were correct.
Miller expected that the boys would do a better job when there was money at stake. Instead, she found that those who were trying to earn the reward made a lot more mistakes than those who weren't. It didn't matter how much they were paid (one cent or fifty cents) or whether they were highly motivated achievers (as measured by a personality test).
...The following year, another graduate student... This time it was undergraduates, 128 of them in all, who were brought into a lab individually. ...As with Millers' experiment, some of the students were informed that they could earn anywhere from $5 to $20--quite a lot of money in 1962--if they succeeded; others weren't promised anything. Even though the subjects were older and the assignment quite different, Glucksberg's results echoed Miller's: when the task was more challenging, those who were working for the financial incentive took nearly 50 percent longer to solve the problem.
...In the early 1970s, a batch of new reports came out that showed the early reports were no flukes. ...It turned out that the children who received candy or the promise of candy got fewer [tests] right than those who received nothing more than information about how well they were doing--a result that led her [Janet Spence] to make the comment... "[Rewards] have effects that interfere with performance in ways that we are only beginning to understand."
Kohn's book is littered with these research stories. Here are a few more good ones:
Each [college student] was asked to work on an interesting spatial-relations puzzle. Half were promised money; the other half weren't. Then the experimenter announced that it would be a few minutes before the next phase of the study got started. The subject was left alone in a room to wait, where he or she could continue playing with the puzzle, read a magazine, or daydream.
Actually, this was the next phase of the study; the subjects were secretly watched to see how long they worked on the puzzle when they had a choice. Those who had been paid... spent less time on it than those who hadn't been paid. It appeared that working for a reward made people less interested in the task. Or, as Deci put it, "money may work to 'buy off' one's intrinsic motivation for an activity."
[Other experimenters (Mark Lepper and his colleagues)] gave fifty-one preschoolers a chance to draw with Magic Markers--something that most children of that age find very appealing. Some of them, however, were told that if they drew pictures they would each receive a special, personalized certificate, decorated with a red ribbon and a gold star. Between a week and two weeks later... those who had been [promised a reward] now seemed to be less interested in drawing with Magic Markers than the other children were--and less interested than they themselves had been before the reward was offered.
It turns out that there are two types of motivation: extrinsic motivation and intrinsic motivation. Extrinsic motivation is the type that comes from without: driven by rewards, praise, punishment, or threats. It's short-lived and fickle. What gets rewarded gets done, yes, but it's done sloppily, cheaply, and always in search of a greater reward. Further, extrinsic motivators tend to destroy whatever intrinsic motivation was already in place.
Intrinsic motivation comes from within. It's what keeps programmers working late into the night, staring at the screen until stomach and bladder demand relief. It's what gives us open-source projects--not the big corporate ones, but the vast majority, the ones created by a programmer or three in their spare time. Intrinsic motivation is powerful and long-lasting.
So, what motivates programmers? Intrinsically motivates them? We've known that for years. Steve McConnell provides a list in his book Rapid Development. The list is a bit old now (sources from 1978 and 1981), but it still rings true. Here are the top ten:
2. Possibility for growth
3. Work itself
4. Personal life
5. Technical-supervision opportunity
7. Interpersonal relations, peers
The key to applying this list is not to try to manipulate your programmers (and other team members) by selecting items to dangle like carrots. Instead, make these things part of your workplace, to the degree that you can, and make them available unconditionally. Then hire people who love to work in the type of environment you have.
"What gets rewarded gets done." It's true... I'm not denying it. As Alfie Kohn says, "Do rewards motivate people? Absolutely. They motivate people to get rewards." But rewards are extrinsic motivators. As an answer to the problem of motivation, they're clear, simple, and wrong. What gets rewarded gets done, yes, but it won't be done well. To get the best results, rely on intrinsic motivation. Find out what intrinsically motivates people on your projects and make that an unconditional part of your work environment.