Why incentives and rewards don’t work

5437288053_624c075aa3_mLet’s talk about incentives. We hear about incentives all the time—especially lately as we try to get the economy going again. We worry about tax incentives. We wonder about the best way to incentivize people to work, or spend money, or save money, or do more work, or better work, or homework.

Being good capitalists, we all know that incentives are great for stimulating all kinds of behavior in the direction that we desire. But are carrots and sticks really effective? And if so, how many carrots are required? Actually, scientists figured this out a long time ago. And I think the answer might surprise you. It surprised me.

But before we get to that, as is my custom, I’d like to share with you a puzzle.

Consider the following problem devised by psychologist Karl Dunker in the 1930’s. It’s called the “candle problem,” and is used for a wide variety of psychological experiments. Imagine sitting at a table when a researcher brings in the following items: a candle, a box full of thumbtacks, and a book of matches. The problem you are assigned is to attach the candle to the wall in a way that prevents the wax from dripping on the table.

Before reading on, think for a moment how you might accomplish this. Many people start by trying to tack the candle to the wall, or lighting a match and melting some wax before attempting to stick it to the wall. Neither works. Eventually, most people stumble upon the solution, which is to empty the box of tacks, tack the box to the wall, and then set the candle in it. The key is in being able to creatively break free from the idea that the box only has one function: to hold the tacks. This block in imagination is called “functional fixedness.”

You’re wondering what this has to do with incentives. So riddle me this: What happens when you give people this sort of problem with incentives for solving it quickly? That’s exactly what Sam Glucksberg, a Princeton University psychologist, wanted to find out. So he timed two groups of people and asked them to solve the candle problem. He told the first group (we’ll call them group A) that he was just timing them to get an idea of how long it normally took people to solve this kind of puzzle. We’re just establishing a benchmark, he told them. No pressure. The second group (group B), he offered incentives. Glucksberg told each participant that if his or her time was among the fastest twenty-five percent of everyone tested, he or she would receive five dollars. If the participant was the fastest, he or she would receive twenty dollars.

I should also mention that Glucksberg did this experiment in 1962. Five and twenty dollars is a nice incentive for many today. Adjusted for inflation, it was an even greater incentive back then. Not bad for a few minutes work.

So you want to know, with all that money on the line, how much faster group B performed? They didn’t. The incentivized group actually took an average of three and a half minutes longer to solve the problem. Three and a half minutes! The monetary reward actually slowed them down. And not by a little!

Experiments of this sort have been performed time and time again. And the results are clear. Traditional incentives don’t always work. And the fact is, they never work when the task requires any amount of creative problem solving. Why? Rewards narrow our focus—which, when a creative solution is wanted, is exactly the wrong thing to do.

I know what you’re thinking: Is it even possible to encourage people work more creatively to solve the complex problems of today’s workplace? It is. But you’ll have to read my next post to learn how. I seem to have lost my motivation to continue with this post.

Founder of WeTeachWeLearn.org, Chris Wondra is just another Wisconsin public school teacher. Email Chris at: mrwondra@weteachwelearn.org.

Image Credit

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>