Monday, October 28, 2013

The Republican Brain Dissected




Chris Mooney's The Republican Brain: The Science of WhyThey Deny Science - and Reality argues that conservatives are inherently more biased than liberals due to psychological differences that are likely partly genetic. To some this may seem almost too obvious to be worth stating. For others, particularly for conservatives, the thesis is liable to provoke a strong reaction. Indeed, Mooney described the book as setting a trap for conservatives. Any criticism of the book’s scientific conclusions can be taken as more evidence for the book’s thesis that conservatives are biased and reject science.

The irony here is that Mooney's thesis - conservatives are inherently more biased – itself runs contrary to the views of many social psychologists working in this area. The current de facto position of social psychologists is summed up by University of California-Irvine psychologist and self-described liberal Peter Ditto (who Mooney quotes in the Republican Brain): “When I’m at home, I spend all my time, like a good liberal, yelling at the television set, denouncing Republicans and how biased they are. Then I assume my persona at work and I say, theoretically, it’s really hard to know why there would be a difference.”


Man is a Rationalizing Animal

That’s not to say that The Republican Brain is all wrong about the connection between politics and personality. Psychological research has found connections between a person’s temperament and their political views, with certain personality types inclined towards conservatism and others towards liberalism. For example, self-described conservatives tend to score higher on personality tests in terms of Conscientiousness, which is characterized by high levels of self-discipline, dutifulness, and organization. Self-described liberals, meanwhile, tend to score higher on Openness to Experience, which is characterized by an inclination towards novelty, variety, and creativity. 

The idea that conservatives and liberals tend to have different temperaments is hardly new. You can find hints of the idea in sources ranging from Thomas Sowell's A Conflict of Visions to ordinary political humor. As with any generalization, there will be exceptions; some liberals will be highly conscientious while some conservatives will be highly experiential. Overall we should not be surprised that conservatives tend to be overrepresented in fields that prize conscientiousness (such as business and the military), while liberals are more drawn to experiential areas such as the arts and the academy. 

In addition, there is a great deal of psychological research establishing that, pace Aristotle, man is not only a rational animal but is also a rationalizing animal, particularly when it comes to politics. When confronted with evidence that goes against their pre-existing beliefs, people do not respond by objectively evaluating it. Instead, they tend to respond emotionally, attacking or dismissing contrary views however possible, while credulously accepting any evidence or authorities that seem to support their view. Anyone who has spent time arguing politics on the Internet will not be surprised at this finding.

Again, these are tendencies, not universals. People can rationally assess contrary arguments, it just doesn’t come naturally. The social psychologist Jonathan Haidt, (whose book The Righteous Mind covers some of the same ground as The Republican Brain but without the partisan overlay) uses the metaphor of a rider sitting atop a giant elephant. The rider is our reason; the elephant, our instincts, emotions, and psychological needs. The rider can move the elephant, but it’s not easy.

Are Conservatives Worse?

Where The Republican Brain departs from Haidt and other social psychologists is in claiming that conservatives are more likely to engage in motivated reasoning in defense of their politics. That’s a much harder case to make, and here Mooney falls short.

Mooney devotes a good part of the book arguing that modern day conservatives are wrong on a whole host of issues ranging from global warming to the effect of Obamacare on the federal deficit. For obvious reasons, these arguments are more likely to appeal to those who share Mooney's liberal politics. 

But even if he were right in every case, this wouldn't be sufficient to justify the claim that conservatives are inherently more biased. To do that, you would have to show conservatives have been more biased and anti-scientific over the whole scope of human history. When one considers left-wing sympathy for pseudo-scientific theories such as Marxist economics, Lysenkoist biology, or Freudian psychoanalysis, there is reason to be skeptical.  


The "asymmetry thesis" also runs contrary to social psychologists' understanding of how motivated reasoning works. Dan Kahan recently summed up this point nicely:
People have a big stake--emotionally and materially--in their standing in affinity groups consisting of individuals of like-minded goals and outlooks. When positions on risks or other policy relevant-facts become symbolically identified with membership in and loyalty to those groups, individuals can thus be expected to engage all manner of information--from empirical data to the credibility of advocates to brute sense impressions--in a manner that aligns their beliefs with the ones that predominate in their group.The kinds of affinity groups that have this sort of significance in people's lives, however, are not confined to "political parties."  People will engage information in a manner that reflects a "myside" bias in connection with their status as students of a particular university and myriad other groups important to their identities.Because these groups aren't either "liberal" or "conservative"--indeed, aren't particularly political at all--it would be odd if this dynamic would manifest itself in an ideologically skewed way in settings in which the relevant groups are ones defined in part by commitment to common political or cultural outlooks.  
To support his thesis, Mooney cites a 2003 meta-analysis of 88 separate surveys comparing personality and political views, conducted by a team of psychologists headed by John Jost of New York University. The Jost study found conservatives were more likely to exhibit dogmatism, intolerance of ambiguity, and general closed-mindedness.

One of the most decisive rejoinders to this cames from Ronald Lindsay, who was also Mooney’s boss (Lindsay is President and CEO of the Center for Inquiry, a non-profit organization that hosted Mooney’s Point of Inquiry podcast). As Lindsay noted, many (though not all) of the underlying studies used in the Jost analysis define conservatism by relying on the “Wilson Patterson Scale,” which categorizes individuals as conservative based on whether they exhibit characteristics such as a “superstitious resistance to science.” Unsurprisingly, if you define someone as conservative if they are anti-science, then it turns out that conservatives tend to be anti-science.    

Mooney’s main argument, though, has to do with the ideological differences in openness to experience. Since liberals are more open to experience, Mooney argues, it stands to reason that they would be more open to contrary points of view, and so will feel less need to use motivated reasoning to swat down evidence that conflicts with their own cherished beliefs.

Failing the Test

It’s an interesting theory, but unfortunately for Mooney, some of the strongest evidence against the theory is set forth in The Republican Brain itself. In the last chapter of The Republican Brain, Mooney describes a study he helped design and carry out along with Everett Young, a psychology post-doc. In the study, students at Louisiana State University were quizzed on their political beliefs as well as their views on a variety of subjects both political (global warming and nuclear power) and non-political (Lady Gaga and Drew Brees). They were then presented with short essays that either supported or criticized their prior views. After reading the essays, the students were asked to rate how persuasive they thought they were, and whether they had changed their minds. [Full disclosure: I went to high school with Drew Brees and think he is a great quarterback.]

Given the human tendency to confirmation bias, one would expect participants not to change their minds much, and to find the essays that supported their views more persuasive than the ones that did not. What set the LSU study apart was that it specifically looked at whether this sort of bias was higher for conservatives on non-political subjects. This is significant, because according to The Republican Brain it is conservatives’ lower openness to experience that is supposedly responsible for their being more biased than liberals. Openness to experience isn't limited to political subjects, so if less openness means more bias, conservatives ought to engage in more motivated reasoning even on subjects that aren’t remotely political.

Mooney clearly intended the study to be his coup de grace, the final, definitive proof that conservatives’ was leading them astray across the broad. There is only one problem. The LSU study failed to find any correlation between political ideology and the tendency to engage in motivated reasoning on non-political subjects. While the book doesn't report findings as to openness, correspondence with Everett Young confirmed that higher openness to experience was not associated with lower motivated reasoning.

If openness to experience doesn’t mean openness to contrary evidence, then Mooney’s whole argument for why conservatives are more biased must fall by the wayside. That, at any rate, is the conclusion his co-author seems to have come to. “My feeling at the conclusion of this study is that motivated reasoning is not a psychological-ideological phenomenon,” says Young. “That is, conservatives and liberals do not differ in their level of motivated reasoning as a result of the inherent psychology that makes them conservative or liberal.”

Admissions like this are devastating to Mooney’s argument. Conservatives criticizing The Republican Brain can be dismissed as utterly predictable. But when Mooney can’t even convince intelligent, liberal-minded folks such as his own co-author or boss (not to mention many liberal social psychologists) that might just be an indication that his arguments aren’t compelling.    

16 comments:

  1. There are some extremely important words missing from this post.

    Like "white." And "south." And "rural." And "class."

    As in, "conservatism" is highly correlated with being white and being from the South and being rural and also class; specifically, being white from anywhere but especially the South (blacks and Latinos in the South are not conservative). Rural people are more likely to be conservative; and the white working class is less conservative than the white upper class, except in the south.

    Here is a survey making racial differences on these kinds of questions clear:

    http://www.americanprogress.org/wp-content/uploads/2013/10/AllInNation-1.pdf

    Also here is a post about how the white working class was basically evenly split between Obama and Romney except in the South:

    http://themonkeycage.org/2012/09/23/puncturing-myths-about-the-white-working-class/

    Point being, if conservatism was really primarily motivated by something inherently genetic, why would it be so systematically linked with region, race, urbanity, and class?

    Does it explain why Asians and Latinos have massively swung towards the Democratic Party in recent years?

    Or could it be that politics is way more complicated than that?

    Certainly there are numerous individual differences that could explain why any given individual might hold the political views they do. But the context for those individual differences are a huge factor.

    ReplyDelete
  2. A very thoughtful review -- I agree that there's not much of a case to be made based on current evidence that conservatives are *inherently* more prone to bias. I think it's worth distinguishing between inherently more biased and more biased based on the current cultural-historico-political climate. Motivated reasoning is not a partisan phenomenon -- I think you illustrated that very nicely -- but the situational causes of motivated reasoning are not necessarily evenly distributed at all times.

    For example, if one party (and I'm not naming any names) demands more allegiance to its core principles than another in order to secure membership, then it follows from Kahan's work that members of that party will be more strongly motivated to reason their way into conformity with the party's platform. If one party has made a big bet on an economic theory that turns out not to be true, then members of that party will have more work to do to twist reality to conform to their pre-existing views. This will make them appear more biased, not because they are inherently more biased but because they have more to be biased about, given the circumstances. Come back twenty years later and the sands may have shifted.

    What's particularly interesting to me, given the current state of the GOP is what useful things psychology has to say about the "fever breaking". When Ziva Kunda wrote her groundbreaking article on motivated reasoning in 1990, she described the idea that people are not free to believe anything they want -- they are constrained by what she called "the illusion of objectivity". That is, once you can no longer plausibly believe what you're trying to convince yourself, the process breaks down. As the great philosopher George Costanza once said, "It's not a lie if you believe it." But once you stop believing it, that's all it is, a lie.

    The question, then, is how far away from reality can you get and keep the illusion of objectivity intact, and what happens when it breaks. Leon Festinger, the father of cognitive dissonance theory wrote a book about a field study he did in the 1950s with a doomsday cult. When their doomsday date came and went without incident, rather than giving up the cult members came to believe even more strongly and proselytized more. But there are only so many doomsdays that can pass non-apocalyptically before even the most ardent believer starts to doubt. The question is when does that happen, and that's a hard one to answer.

    For example, the strength of forces working in the opposite direction can vary greatly. The cult members had sold off their possessions -- they didn't have much to go back to if they admitted they were wrong. In the GOP, allegiance to the group is strong and the cost of speaking out can be high; witness the "moderates" who have been excommunicated or primaried. The relative power of the "true believers" and the "pragmatists" is also very much in question. To necessarily oversimplify, the fiscal conservatives used the ardor of social conservatives to great profit, but they may have lost control over the process in the process and now that their profits are in the hands of prophets it may be hard to wrest the power back. Ok, now I'm just rambling and amusing myself with silly wordplay, but it's a really interesting question and we have only the vaguest answers.

    Cheers, and thanks for a very thoughtful piece.

    ReplyDelete
    Replies
    1. Nathanael7:55 AM

      Thanks for a genuinely thoughtful and well-informed piece, Mr. Nussbaum. I didn't think Mr. Neeley's piece was worth a warm bucket of spit, but yours is genuinely useful.

      Delete
  3. Shenanigans. What brain.

    ReplyDelete
  4. The problem with Chris Mooney is that he knows quite well that the big science-politics issues of the 21st Century revolve around IQ, race, and sex, and that liberals are more actively anti-science in those areas. James D. Watson and Larry Summers weren't forced into resignation from Cold Springs Harbor Laboratory and Harvard, respectively, by the Tea Party.

    ReplyDelete
    Replies
    1. Anonymous2:53 PM

      Say what ?

      Delete
    2. Anonymous3:58 PM

      Sailer is a one trick pony. IQ is fake science designed to support status quo conservative concepts of race and sex and he is chin deep committed to these 19th century racist models that support good old fashioned conservative values. There will be no questioning of the measuring instrument because science!11! Watson ripped off the double helix model from a woman, and Summers is simply a pig in all ways; here Sailer commits the trifecta of conservative reason: argument from authority, argument from victimhood, and argument from common sense.

      Delete
  5. Um, not so fast:

    http://www.cell.com/current-biology/abstract/S0960-9822%2811%2900289-2

    Poor babies, it might not be their fault.

    ReplyDelete
  6. "pseudo-scientific theories such as Marxist economics"

    Just leave out the Marxist bit and it is still correct.

    ReplyDelete
  7. I think there's a fundamental flaw in this review (and perhaps in the book itself, although I have not read it) - the risk of taking labels at face value is that you may fall victim to Inigo's Observation. It may not be as classic a blunder as going in against a Sicilian when DEATH is on the line, but it's pretty classic nonetheless. I would contend that considering modern, self-identified, Conservatives to be the same as 'small-c' conservatives is a, as The Spleen would say, Big Mistake. The people who presently call themselves Conservatives are largely, I contend, not conservative at all, but very radical reactionaries. It's as if you're trying to define "liberals" using the qualities of Trotskyite or Shining Path Marxists (as if anyone would be stupid enough to do that!). Other than that small flaw, there seems to be a good bit of truth to the basic idea. Unfortunately that all boils down to "We're human; what can you do?" I'd have to look at the Mooney study you cite, but from your description your observation doesn't seem unlikely - the problem with social science studies in general is that all too often the effort of forcing human variables onto the Procrustean bed of numbers and simple categories results in something which looks meaningful but may be a lot more murky than desired (or vice-versa). It's hard enough in the so-called "hard sciences" to accurately extract meaning from data; I tend to be skeptical of broad conclusions drawn from sociological studies done on small pools of college-age students. Be very leery of labels and sweeping conclusions, especially when there may be alternative explanations which have been ignored in the interests of more appealing apparent coherence..

    ReplyDelete
  8. Anonymous2:02 PM

    A much more interesting study would rank people on degree of political partisanship (Democrat or Republican or whatever) and see whether partisanship itself is an indicator of motivated reasoning and bias both within and outside of the political sphere.

    For example, we could casually introduce a non-political paper as having been written by a Republican and see whether high partisanship people process information differently from low partisanship people.

    I strongly suspect that this is the real axis that determines emotional override vs. reason / rationality. I've seen it in action so much and so consistently.

    ReplyDelete
  9. I think Dave Nussbaum is on to the right track here. Conservatives aren't *inherently* any more given to rationalization than liberals. A liberal myself, I know too many anti-scientific liberals to fall for that allegation. (GMOs, anyone?) The problem is, *current* conservative ideology is *very* much further removed from reality than liberal ideology. Conservatives just have more to rationalize. And boy, do they. ("Reality has a well known liberal bias."--Stephen Colbert)

    ReplyDelete
  10. Anonymous5:34 PM

    The answer is pretty obvious. Mooney wanted to sell books. Books that call Republicans Evil Nazis sell well. My aunt has entire shelves of them, none of which she has read.

    ReplyDelete
  11. Nathanael7:48 AM

    Sorry, but Mooney is right. Political conservatives ARE worse. The evidence for this is overwhelming.

    Now, why is this? I think it's due to the definition of "political conservative" in this day and age -- as an earlier commenter said, the modern "political conservative" is actually not conservative at all, but a radical extremist reactionary ideologue.

    All the sensible conservatives are now called "not really conservative" and thrown out of the Republican Party, and even out of the "conservative" third parties.

    At different times and places in history, the people referred to as "left-wingers" politically may have been the nutsoid ones -- this might have been true in 1918 Russia, for example -- but at this time and place in history in the US in 2013, the *words have been redefined* so that "conservative" refers specifically to people who are especially badly biased.

    In short, Mooney's right, but the reason why he's right is linguistic and therefore not terribly interesting.

    "Dictionary" conservatives -- people who fit the dictionary definition of "conservative" -- are not any more biased than average, but we're also not considered "political conservatives" any more.

    ReplyDelete
  12. Nathanael7:53 AM

    ". In the study, students at Louisiana State University were quizzed on their political beliefs as well as their views on a variety of subjects both political (global warming and nuclear power) and non-political (Lady Gaga and Drew Brees). They were then presented with short essays that either supported or criticized their prior views. After reading the essays, the students were asked to rate how persuasive they thought they were, and whether they had changed their minds."

    You've made a mistake here: global warming is not political, and neither is nuclear power. However, Lady Gaga is extremely political and frankly so is Drew Brees.

    You don't have a functional definition of political.

    The difference is that the Republican Party bosses have not recognized that Lady Gaga is an issue and have therefore not started ordering their followers to hate her. The conclusion is that political conservatives engage in biased thinking when they've been ordered to by their party bosses. When they haven't been given any orders... they have no prior biases to work from.

    In short, political conservatives are goose-stepping authoritarian bots. Political liberals are not, or at least not necessarily (I'm sure some are).

    ReplyDelete
  13. Anonymous11:13 PM

    The problem here is that the distinction was made in terms of political ideology.

    To see why this is a problem, see:

    http://home.cc.umanitoba.ca/%7Ealtemey/

    This is Bob Altemeyer's The Authoritarians that looks at the authoritarian personality. Altemeyer is probably the world's leading authority on this question. As can be seen in the book, the question isn't dependent on political ideology, but on authoritarian traits. These turn up in people with all different kinds of political views; it has more to do with how they react to outgroups and to authority figures then anything else.

    So Kuban is roughly right; motivated reasoning has different roots then political ideology does. But … one of Altemeyer's findings is that, in fact, one is more likely to find authoritarians among conservatives then among liberals, at least in the US. This shouldn't be that surprising, given the disproportionate role of authoritarian leaders in their world. Still, it pays to keep an eye out for these people all over the political spectrum.

    ReplyDelete