Beatles versus Rolling Stones. Ironman versus the Incredible Hulk. Deep dish versus thin crust. Such differences of opinion among family and friends rarely end in serious squabbles. Let the conversation turn to political parties, however, and lively disagreements can become downright ugly.
Why is it that even among the people we care about most, differences in political affiliation often result in awkwardness and discomfort, and pushed far enough, can feel like a threat to the entire relationship?
The answer may lie in research conducted at UC Santa Barbara's Center for Evolutionary Psychology, where social scientists sought to understand how and why the human brain -- below the level of conscious awareness -- categorizes political parties. "We found that differences in political opinions engage the brain's evolved circuitry for tracking alliances and coalitions," said David Pietraszewski, a postdoctoral fellow at the Max Planck Institute in Germany and lead author of a paper published online in the journal Cognition.
"When people express opinions that reflect the views of different political parties, our minds automatically and spontaneously assign them to rival coalitions," he continued. "As far as our brains are concerned, political affiliation is viewed more like membership in a gang or clique than as a dispassionate philosophical stance." Think biker gang, not debate club.
What's more, as this evolved system notes and retrieves information about an individual's political alliances, it begins to ignore other possible cues about who is allied with whom. And one of those cues it ignores is race. "This decline in the mind's tendency to categorize people by their race happens when race does not predict alliances, but other cues do," said Pietraszewski. "It's a telltale sign that our minds are treating political opinions as markers of membership in a coalition."
"Our brains are not designed to attend to race," explained John Tooby, professor of anthropology at UCSB, co-director of the Center for Evolutionary Psychology and an author of the paper. "Instead, they are designed to attend to coalition -- and race gets picked up only as long as it predicts who is allied with whom. This is why successful politicians like Benjamin Disraeli, Arnold Schwarzenegger or Barack Obama need not be ethnically the same as the majority of their supporters. Coalition is the real coin of the evolved mind, not race."
Humans come from an evolutionary history that included conflict among groups or factions, Tooby added, and it was important for individuals to know, if a dispute were to break out, which individuals line up with "us" and which with "them." "While the world is full of social categories like athletes, plumbers, the elderly or nail-biters, only a few categories are interpreted by the mind as coalitions -- sets of individuals inclined to act together, and support each other against rivals," he said. "In the small social world of our ancestors, the political was personal."
For our hunter-gatherer ancestors, guessing incorrectly about who is allied with whom would have had very real consequences, noted Leda Cosmides, UCSB professor of psychology, co-director of the Center for Evolutionary Psychology and also an author of the paper. "This is why we hypothesized that natural selection designed the brain to automatically construct social maps of local coalitions out of clues that imply or predict alliance," she said.
To test their hypothesis that political affiliation non-consciously triggers the mind's "us versus them" system, the researchers showed participants a calm and civilized discussion between eight Republicans and Democrats. Each side was composed of two black and two white individuals, and all espoused opinions typical for their respective parties. Participants were then shown excerpts from the conversation and were asked to indicate which individual expressed each opinion. The results showed that participants spontaneously categorized speakers by their political party, and this caused a decrease in racial categorization.
"Because we live in a society where race predicts patterns of mutual support -- of cooperation and conflict -- our mind's alliance detection system spontaneously assigns people to racial groups and uses those categories when there are not other clues to alliances," explained Cosmides. "For years, psychologists tried many different ways to reduce racial categorization, but all of them failed. They thought it might be irreversible. But prior research at our center showed that there is one social context that easily and reliably decreases racial categorization. When race no longer predicts coalitional alliances, but other cues do, the tendency to nonconsciously treat individuals as members of racial categories fades, and sometimes disappears."
The researchers' previous work, Pietreszewski pointed out, shows that this effect is specific to alliance categories. "Coalition membership has no effect on categorization by gender -- and now we know it has no effect of categorization by age either," he said.
The researchers conducted parallel experiments varying gender or age instead of race. In the sex experiments, each political party was composed of two young men and two young women. In the age experiments, each party was composed of two 20 year olds and two 70 year olds (all of the same gender). Participants strongly categorized the speakers by their political party, whether their members varied in race, gender or age. When they did, racial categorization decreased, but categorization by gender and age remained high -- indeed, just as high as when no information about party membership was provided.
"Categorizing people as Republicans versus Democrats caused a decline in categorization by race, but not by gender or age," Pietraszewski said. "This is what you would expect if the mind treats race as an alliance category."
"Our minds spontaneously categorize people as male or female, young or old," Cosmides explained. "These are fundamental social categories: they organized the social life of our hunter-gatherer ancestors across many different social contexts -- mating, parenting, hunting, gathering, and warfare, to name a few. Yes, alliances based on differences in gender or in age sometimes exist. But many different mechanisms in the mind need to know this information. For this reason, the circuitry that records and retrieves people's gender and age should operate independently of the alliance detection system."
According to Pietraszewski, this pattern -- categorization by political party decreasing categorization by race, but not gender or age -- was predicted ahead of time. "It follows from the hypothesis that our minds treat both race and politics as alliance cues," he said.
This explains the heated discussions and often-uncomfortable barbs that arise when holiday dinner conversations veer into political waters. "They are not a dispassionate consideration of alternative views," said Cosmides. "The views are flags planted, marking your coalitional alliances."
The bad news is that once constructed, it's easy for our minds to frame alliance categories like race and politics in terms of an "us versus them" mentality, the researchers explained. But the good news is that these results show that race and politics are intrinsically flexible categories as far as our minds are concerned. "Our previous research -- and our politics study -- show that it is not impossible to change these "us versus them" perceptions, even for something like race," Pietraszewski said. "What is required is cooperation that cross-cuts the previous boundary, and the more the better. Reducing racial discrimination or political polarization will be no easier or harder than changing patterns of cooperation.
"The experimental work shows that it is possible to make these divisions fade," he continued. "How to make this happen is not a mystery anymore."
Other co-authors of the paper include Oliver Scott Curry of the Institute for Cognitive and Evolutionary Anthropology at Oxford University and Michael Bang Petersen of the Department of Political Science at Aarhus University.