Psyched!

I was typing something up on these same studies when I saw this.
We sure do write on the exact same stuff a whole lot.

Anyway, I think it has to do with self-presentation/not wanting to look like an idiot/whatever else way of putting it. It seems to me that people in the good luck conditions would be more motivated to do well than those in the other groups.

In real life this heightened motivation could perhaps backfire. But in a lab study like this, the non good luck charm group could probably give 2flying craps about how well they do. The good luck charm group simply cares more. And sense the tasks arent that hard, that is enough (just paying attention etc).

If you brought in your good luck charm and failed you would really feel like a dumbass. Thats all I am saying, and it could explain these results.

A way to test if this is behind them would be to use more difficult tasks, or up the incentive to do well on the tasks.

I have a quirky superstition that if I use product placement in an essay, people will think less of me and turn away in disgust.

Confidence is important in performance, too much or too little and the performance will be lower than the average performance. Based on confidence, the lucky object/superstition will work for them in the same way as just confidence for people who don't have an object or superstition, won't it?

I wear a different "lucky" perfume everytime I take a test or need a little luck in my life. However, I find that it seems to work only half of the time.

This is an interesting experiment. Luck while gambling is a myth, however, there may be some sort of mental mechanism that helps a person perform better when they feel lucky (and when the task depends on their performance, unlike e.g. blackjack).

I highly encourage the experimenter to repeat these tests with a much higher population. You should be trying for 5-6 sigma levels of confidence, in order to be extra, extra sure these results aren't due to chance!