This might be the best example of unconscious and implicit priming yet. This study concluded that hurricanes with female names cause more damage and more deaths than hurricanes with male names.
Of course, as you might suspect, it can't be the gender. Hurricanes get male and female names on alternate years. And the gender is not really related to the hurricane. So what can it be?
The researchers suspect that it is because of our unconscious mental models of gender. When we see on the news that hurricane Christina is bearing down on us, the unconscious and visceral feeling we get is not as great as for hurricane Christopher. So we are less likely to take precautions - and are thereby more likely to experience damage or injury.
But Shankar Vedantam (of "Hidden Brain" fame and one of my favorite experts regularly appearing on NPR) pointed me to this article that calls into question the main finding, explaining away the data with some methodological issues in the original study.
But I hope that the study is correct. Not because I want us to have unconscious gender stereotypes but because unconscious priming is such as fascinating phenomenon. It pops up over and over again. This is one of my favorites. So perhaps I am suffering from a confirmation bias of my own - hoping that it is true makes me more likely to believe it and to discount the contradictory evidence.