Responding to Incentives Sobering Up

We know that people respond to incentives, if for no other reason than that our models based on this assumption do a pretty decent job of explaining and predicting reality. Sometimes (surprise!) people even respond to monetary incentives:

Social drinking can be fun, but sometimes knocking back just a few
can get a person in a lot of trouble. Alcohol is a depressant, thereby weakening the control we have over our behaviour and making us less inhibited. After just a few drinks we may tell off our boss at the Christmas party or flirt shamelessly with someone. Aha, say some people, as long as the drinking isn't too excessive, these problems can be licked with either strong coffee or a personal incentive to stay out of trouble. How true is this folk wisdom?

Research psychologists at the University of Waterloo got a bunch of
undergraduates sloshed to find the answer. The researchers were interested in how much control could be regained by drunk students if they were given either caffeine (equal to about 2 1/2 cups of coffee) or a small financial incentive to sober up. Upon being inebriated with an alcohol-laced soft drink (equivalent to about three beers for someone weighing 150 pounds), the students were asked to complete a rather complex word game. The students who had been given just alcohol performed the worst, the drunk students with the small financial incentive to do well in the word game performed the best, and those who were given caffeine performed somewhere in the middle.

The study, appearing in this month's Experimental and Clinical Psychopharmacology journal, only looked at behavioural control and has no implications for other effects of alcohol, such as diminished reflexes and spatial distortion.

My worry is that some well-meaning politician will now argue we should pay people for being sober when they drive.

I wonder if students had to be paid or were willing to pay to be involved in this experiment. [h/t to BF]