"If the payments are too high, they may backfire," said paper co-author Nina Mazar, a marketing professor at the University of Toronto's Rotman School of Management. "I don't think people realize that there is a threshold."

Experiments conducted in India and the United States showed that participants doing tasks requiring creative thinking, problem-solving or memory skills improved their performance when financial incentives were increased to moderate levels. But once the incentives went beyond a certain threshold, they "choked" and did not perform as well.

"What we were really surprised by is that it didn't matter what kind of task -- as soon as it involved even a little bit of cognition, very high incentives did not work," Mazar said.

But the findings were not consistent for tasks that were purely rote and mechanical. In that case, performance contingent payments "seem to work very well."

Participants were variously given games to play, anagrams to solve, adding tasks and jobs involving pressing keys. Researchers chose to run one experiment in rural India because it gave them the chance to offer relatively high financial incentives (up to half a year's income) on a modest research budget. Another experiment used American university students at the end of an academic term, when they would be more likely to be running low on funds and therefore be more responsive to financial incentives.

Mazar said there was "no difference" in the study's findings among the different experiment settings.

"There is evidence from psychology that too high levels of stimulation can harm performance," she said. "We show that a similar relationship can be found for financial incentives and performance."

The complete study is available at http://www.rotman.utoronto.ca/newthinking/largestakes.pdf .

Provided by University of Toronto