Stanford promised a certificate and the children drew less

In 1973, Mark Lepper, David Greene and Richard Nisbett, three psychologists at Stanford University, walked into a nursery classroom with an odd requirement: they only wanted children who ALREADY liked drawing. They spent several days watching free play through a mirror and kept the ones who picked up the felt-tips without being asked.

There were fifty-one three-to-five-year-olds from the university’s own nursery school. They split them into three groups. The first was shown, before starting, a certificate with a ribbon and their name on it, and told they would get it if they drew. The second was promised nothing and, when they finished, was handed the same certificate as a surprise. The third got neither prize nor promise: felt-tips and that was that. Each child drew alone for a short while, in a separate room, with an adult making conversation.

The certificate was a piece of card with a gold star, a red ribbon and the words Good Player.

A week or two later the researchers came back to the classroom, put the felt-tips out on the table on an ordinary day, with no adults visibly taking notes, and timed them. The surprise-prize children and the no-prize children drew as much as ever. The children who had been promised a certificate drew for about half as long. The measurement was taken like that, quietly and in their usual classroom, so that no child would connect those felt-tips with a certificate from a fortnight earlier.

Nothing had been taken away from them, and nobody had told them off. They had simply been told once, very nicely, that drawing was a way to get a piece of card with a ribbon on it.

And the surprise-prize children had been given exactly the same card with exactly the same ribbon. The only thing that changed was when they were told about it.

This is called the overjustification effect, and what it says is narrow: if someone already does a thing for pleasure, putting a price on that thing is the fastest way to find out what the pleasure was worth. It is not about wages, or about hobbies in general.

Fifty years of replications have narrowed the effect down to one condition you can recognise on sight: the reward has to be promised in advance and has to sound like control (“if you do this, I’ll give you that”). Specific praise for what somebody has just done breaks nothing. The most cited review of all — a 1999 meta-analysis by Edward Deci, Richard Koestner and Richard Ryan pooling 128 experiments — stopped at that same border: tangible, expected rewards sink intrinsic motivation, and verbal praise does not. Carried over to an office — and this is my reading, not a laboratory result — a bonus on every fault fixed does get faults fixed, and along the way it puts a price on something that until that day was being done for nothing.

Translated into tomorrow morning:

1. Before you offer anything, check whether it is already happening. If your child opens books unprompted, if a colleague fixes things in the workshop because it amuses him, if you play the guitar on Sundays: that already works without you putting money on top of it.

2. Ask about what went on inside, not about the scoreboard. “Which bit surprised you?” instead of “how many pages have you done?”. “How did you track the fault down?” instead of “how many did you close today?”.

3. If you want to celebrate it, celebrate afterwards and without having announced it. Dinner because the book got finished is fine; dinner promised for every book is not. The difference is that the second one was agreed in advance.

4. Save agreed rewards for the things nobody would do for pleasure: sorting an archive, covering the Saturday shift, getting paperwork done. There is no motivation to protect there and the deal is an honest one.

5. If you have already built a points system and want to dismantle it, do not withdraw it quietly or overnight: pay what you promised, explain that the deal was turning something enjoyable into an obligation, and stop announcing the next one.

And one last one, the cheapest of the lot: when you catch someone doing for pleasure something that also happens to suit you, do not promise them anything in exchange for carrying on. Tell them afterwards what you liked, if you want, but afterwards and with no deal attached.