It's probably a big disappointment to the national press corps, which has been getting a lot of mileage out of the "race card" lately, that Hillary Clinton and Barack Obama have moved on to whether or not a messy desk is a good thing. In case you missed the Nevada debate, Obama revealed he's not good with organization, a fatal flaw for a future president, according to Clinton.

Barack, I feel your pain. My desk is always a disaster. I used to be embarrassed about this until I saw a picture of Albert Einstein's desk and read a book called A Perfect Mess, which explains why people with messy desks are in every way better than people with neat desks. I am guessing the Clinton staffers haven't seen the photo or read the book, or they would realize they have stumbled into a minefield from which they should extricate themselves, but quick.

Alas, Clinton has proven how out of touch she is with ordinary folks (statistically most people have messy desks) by coming down on the side of the neatniks. She has jumped on her opponent's lack of office management skills to prove that he's another George W. Bush - a vision-thing and affability puppet who is going to conveniently lose important papers before they can be subpoenaed. She, on the other hand, would cross-shred and incinerate hers.

The whole neat vs. messy desk debate only reaffirms my method for choosing a presidential candidate: gut instinct. My policy-wonk friends - who watched every debate, who actually read position statements, who can tell you the difference, in detail, between Clinton's health care plans and Obama's - are horrified by this. But the truth is, however intellectual and objective they believe their decision-making was, they were, at least in part, voting on gut instinct, too.

Social scientists, psychologists and neurologists have long studied how subconscious beliefs shape our actions. They've got all kinds of clever tests and brain scans that indicate pretty strongly that there's often a disconnect between our implicit (unconscious) attitudes and our explicit (stated) attitudes.

Now you'd think that we'd all

have a general bias toward people who are like ourselves. But Harvard University's Project Implicit, which has gathered data from more than 4.5 million people in the past 10 years, has revealed that some group preferences are pretty counterintuitive.

For example, in tests for racial preference, it's not surprising that 75-80 percent of whites show a preference for whites over blacks. What is amazing, though, is that 50 percent of black test-takers show a preference for whites over blacks, a result, the researchers theorize, that may be caused by the way blacks and whites are portrayed in popular culture. The same holds true for age - 80 percent of respondents show negative feelings toward the elderly, including a majority of people over age 60.

As long as we're making front-of-the-brain, higher thinking decisions, the kind where we take our time and consider lots of options, it's possible to overcome implicit biases. But when we have to make a spur-of-the-moment decision, for example if you're a still undecided voter standing in the voting booth with a ballot in your hand, then those biases can come into play.

Luckily, though, we've got a jillion of those biases running around in our brains, clicking into place, creating one great decision-making algorithm. And as often as not, our decision will be based on a quick first impression of a candidate, what cognitive scientists call "thin slicing" an ability defined by Malcolm Gladwell in his book Blink as unconsciously finding "patterns in situations and behavior based on very narrow slices of experience."

Thin slicing is an ability that has kept humankind alive for millions of years. It goes way beyond judging an individual by sex or race to incorporate hundreds of subtle clues given off by body language, tone of voice, word choice, etc. We instinctively like and dislike people, trust them and don't trust them. We are not always correct in these instincts, but they are very powerful and hard to shake. (next page »)