Scientists say that James Carville was only partly right when he famously quipped: It's the economy, stupid. A new study adds weight to the idea that it's also genes that sway our choices in the voting booth.

While political commentators were quick to pin last week's midterm Republican victories on everything from the tanking economy to the president's agenda, plenty of people still pulled the lever for Democrats. Deep-seated liberal and conservative philosophies still drive our voting behavior, and scientists are trying to figure out how much of that is attributable to nature (our genes) and how much to nurture (the environment).

Until a few years ago, upbringing, experience and culture were considered the primary factors influencing our political preferences. According to such thinking, if Sarah Palin were raised by liberals in a big city and Nancy Pelosi had grown up hunting moose in Alaska, they'd be giving each others' speeches.

But that thinking is changing. This most recent study, published in the October issue of the Journal of Politics, revealed a complicated connection among political preferences, a gene called DRD4, and even the number of friends that people said they had in high school.

The DRD4 gene influences the brain chemical dopamine, which is involved in movement, emotion, pleasure, and pain.

About 38 percent of us carry an alternative version of the gene which has been associated with a novelty-seeking trait. Five percent of us carry two copies of the novelty-seeking version.

"A novelty-seeking individual would be more likely to try a new kind of food or climb Mount Everest," said James Fowler, a professor of political science and medical genetics at the University of California San Diego. That's what made him think it might have some connection to politics.

Fowler, who is lead author on this latest study, said the findings have already been misinterpreted, thanks to the polarized political climate we live in. "The conservative blogosphere has really picked up the story as a finding of a gene for some disease - a gene for liberalism," he said.

But the way he sees it, these variants in our DRD4 genes have been around for tens of thousands of years. If one form were superior, the other would have been weeded out by natural selection.

Fowler's study builds on an earlier one published in 2005, showing the first concrete connection between genes and political attitudes.

John Hibbing of the University of Nebraska and colleagues studied thousands of twins and found that identical twins were more likely to share political attitudes than were fraternal twins.

Comparing identical and fraternal twins has become a standard way to measure a general genetic component or "heritability" of a trait. Both types of twins share very similar environments from the womb onward, but identical twins share all of their genetic information, while fraternal twins are same-age siblings, sharing about half of their genes.

Such twin studies have shown that autism, homosexuality, and a number of other traits stem from a combination of genetic and environmental factors.

Hibbing said he measured political preference with a standard battery of questions on issues such as abortion, guns, and gay marriage. In the last several years, similar studies in Australia and Denmark have shown a genetic component to those beliefs.

But those studies couldn't point to any particular gene.

The new results don't directly connect voting with the DRD4 gene - the relationship is mediated by the environment.

Still, among those who carry the novelty-seeking variant of the DRD4 gene, the researchers found a connection between self-reported liberalism and the number of friends in high school. Those with more friends were more liberal.

Among those with the more common version of the gene, there was no association between the number of high school buddies and political persuasion.

Fowler said one possible explanation is that these novelty-seeking people would be more likely to explore the viewpoints of their friends on social, religious and political issues. All that exploration would tilt them toward the liberal side of the spectrum.

Those with fewer friends would, in theory, lack enough alternative views to explore.

Even the authors of the study suggest caution in over-interpreting the results. The significance of these variations in DRD4 remains in some dispute, said Joel Gelernter, a psychiatrist at Yale University. "The novelty-seeking stuff hasn't held up very well but to my knowledge hasn't been looked at carefully," he said. The evidence is stronger that it's connected with attention deficit disorder.

In the meantime, Nebraska's Hibbing, who did the twin study, said he had begun studying physiological differences between liberals and conservatives. He's found that conservatives are more likely to react strongly to threatening images - something he measured through skin conductance, a standard way to measure bodily response. He said they also have faster eyeblink reflexes, which he thinks might be related to a conservative affinity for policies that promote safety from communists and terrorists.

"When conservatives hear this, they think I'm calling them chickens," he said.

And yet, liberals are more strongly associated with a cautious stand on pollution, global warming, and potential carcinogens in consumer products. Hibbing acknowledged that the division is a complicated one.

He said he hoped that eventually connecting biology with political views would promote tolerance. "These are different ways of approaching life," he said. "People are not being willfully bullheaded."