This post was initially published on September 16, 2004. It takes a critical look at some UCLA studies on brain responses of partisan voters exposed to images of Bush and Kerry:
Using M.R.I.’s to See Politics on the Brain
The researchers do not claim to have figured out either party’s brain yet, since they have not finished this experiment. But they have already noticed intriguing patterns in how Democrats and Republicans look at candidates. They have tested 11 subjects and say they need to test twice that many to confirm the trend.
Do liberals ”think” with their limbic system more than conservatives do? As it happens, some early research suggests that Armey might have been on to something after all. As The Times reported not long ago, a team of U.C.L.A. researchers analyzed the neural activity of Republicans and Democrats as they viewed a series of images from campaign ads. And the early data suggested that the most salient predictor of a ”Democrat brain” was amygdala activity responding to certain images of violence: either the Bush ads that featured shots of a smoldering ground zero or the famous ”Daisy” ad from Lyndon B. Johnson’s 1964 campaign that ends with a mushroom cloud. Such brain
activity indicates a kind of gut response, operating below the level of conscious control.
On Tuesday, the New York Times ran a front-page story about a study that analyzed the brains of political partisans as they viewed Kerry and Bush campaign ads. If you thought Googling for yourself was the sign of a self-obsessed culture, get ready for the personal brain scan.
Is there something intrinsically reductive or fatalistic in
connecting political values to brain functioning? No more so than ascribing them to race or economic background, which we happily do without second thought. Isn’t it more dehumanizing to attribute your beliefs to economic conditions outside your control? At least your brain is inalienably yours — it’s where the whole category ”you” originates. No one denies that social conditions shape political values. But the link between the brain and the polis is still uncharted terrain. Prozac showed us that the slightest tinkering with brain chemistry could have transformative effects on a person’s worldview. Who is to say those effects don’t travel all the way to the voting booth?
I’ve been thinking about the UCLA study ever since that first NYT article came out, and my thoughts went in two different directions. First, I was wondering how the experimental protocol can be improved to make the data more useful. Second, I was wondering about the possible interpretations of the data. Let me try to put on cyberpaper some of my trains of thought and see for myself if any of that even makes sense.
1) How to make a better protocol. Brain scans are very, very expensive. Is there a cheaper way to do this that would allow for a much greater sample size to be used in the study?
What their preliminary data suggest is that some people (Democrats) respond to some images (mushroom clowd or Twin Towers burning) with greater activity in the amygdala. All that says is that some people respond to these images more emotionally than others. It is not neccessary to watch the amygdala in order to see if someone has an emotional response to an image. While a polygraph is lousy at detecting lies, it is great at detecting emotional responses, and is a million times cheaper, not to mention the ease of use.
What both techniques lack is the ability to distinguish between different types of emotions, e.g., fear, love, etc. I think that it is neccessary to introduce a separate way of getting at that kind of information and, unfortunately, self-reporting seems to be the only way. Speaking out the “name” of the emotion as it happens, pushing a button or turning a dial on a simple apparatus, or filling a questionnaire immediately after the test, may be some ways to get at this.
So, the UCLA people showed their subjects pictures of Bush, Kerry and Nader (no emotional response), followed by videos of either “Daisy” or “9/11″ ads (elicited emotional response in Democrats only), followed by second run at the pictures of the three candidates (Democrats responding emotionally to Kerry, Republicans to Bush). We are assuming that both videos induced negative emotions of fear or anger. I would add some controls here, e.g., including images that elicit a positive emotional response.
Also, both videos show events that have been highly politicized. I would add an image that elicits a negative emotion but is not politicized, e.g., devastation by a hurricane. Finally, the “Daisy” ad, although produced by a Democrat against a Republican (as opposed to the “9/11″ ad – produced by Bush against Kerry) is so old that it may not have much of an effect even on the people who may remember it and remember who was who in that election, i.e., it has depoliticized itself over time for at least some of the subjects.
Another problem I have is with showing images of Bush, Kerry and Nader. There are many “likeability” polls out there that suggest that Republicans like Bush much more than Democrats like Kerry. This may skew the data somewhat because what is measured is the emotional response. I would level the playing field by showing pictures of WHOLE CABINETS, including the most recognizible faces from the two parties. For instance one image shows Bush, Cheney, Rumsfeld, Aschcroft, Rice, Wolfowitz, and the other one shows Kerry, Edwards, Clinton, Carter, Gore, Dean and Hillary. Forget Nader – he is a non-factor here (although his picture does serve as a kind of control). Alernatively, one can show symbols instead of faces, e.g., the recognizable logos of elephant and donkey.
Finally, where are the Independents/Undecideds/Moderates/Inerts? Aren’t they the key question of the study? Isn’t it what their brains do what the campaign managers really want to know? Is this a way to identify who will vote for whom in event of an October surprise, for instance?
2) Interpretation of data. I am happy to see that nobody, so far, suggested the existence of a “conservative gene” and a “liberal gene”. Actually, not much interpretation has been offered at all, beyond pointing to the results: Democrats tend to respond more emotionally to scenes of devastation; and, both Democrats and Republicans respond emotionally to the image of their leader if shown immediatelly after the images of devastation. One article mentions that the patterns of brain activity are formed by experience and stops at that.
What kind of experience can affect the brain patterns in ways shown in the study? What accounts for such sharp differences between conservatives and liberals? I will assume, by my own experience, that people thinking about this question will inevitably mention learning. I will suggest that this is a wrong way to think about it.
Model of learning most often trotted out is the learning of courtship song in songbirds. Father-bird is the tutor and son-bird is the student. After some listening and some practicing, the son-bird learns how to perfectly repeat his father’s song. Thus, the local (or even familial) song gets transmitted from one generation to the next and the next and the next. This model implies that conservative and liberal worldviews are directly taught (by parents, naighbors, friends, teachers, preachers, preachers’ sons, media etc.).
I prefer a different model. There was a study a few years back in either Science of Nature (I had trouble finding it in the database this morning, but I remember it well) of some kind of songbird (a sparrow of some kind?). Males were either sporting shiny blue feathers on the chest (preferred by females) or were drab-colored. The feather coloration was not determined by genes in any way. It was determined by the bird’s overal health. The health was affected by the parasite load. The size of the parasite load was the result of growing up in a clean vs. dirty nest. In the division of labor of this species, it is the job of the father to clean the nest (by eating the bugs and worms in it). Healthy males with blue feathers are superb nest-cleaners and their sons, thus, grow up healthy, develop blue feathers and – clean their nests better! What the females are looking for are blue feathers. What is really transmitted from one generation to another is the habit of cleaning the nest really well. The study did not fully adress the mode of transmittion of the ability to clean the nest. It is likely that the sons learn the cleaning TECHNIQUE by watching their fathers (direct learning), but it is much more likely that they inherit the NEED to clean the nest (or to antropomorphize, the understanding of the importance of a clean nest) not by learning but by growing up in a clean nest. It is not watching their father, but growing up in a clean nest that makes the son a good or a bad cleaner. The father transmits the “value” of the good cleaning by providing a clean nest. It is not direct learning, but growing up in a particular environment that is the mechanism of intergenerational transmition of the key trait (blue feathers).
Back to politics. I do not believe that conservative or liberal values are taught. I believe they are the result of growing up in a particular environment, characterized by a particular childrearing philosophy of the parents. I wrote more about this below, in the “Moral Order” post. The conservative Strict Father model of childrearing is a harsh stick-and-carrot approach. If a child falls and scrapes his/her knees, the parent adds the insult to the injury by chiding the child for not paying attention or for being clumsy. The point is to make the child “tough” and self-reliant. A Nurturant Parent will pick up the kid and give hugs and kisses, fully understanding that falling is a natural part of growing up, and that the experience of falling naturally leads to being more careful (and nimble) in the future. From a child’s perspective, sudden pain and fear is aggravated by the beloved parent getting angry. The child withdraws and learns to control one’s own emotions. The child becomes numb to its own feelings and emotions (almost autistic in a sense). A person who does not feel much about oneself cannot possibly feel empathy for other people. Thus a Republican is raised. Of course there will be no lighting up of amygdala at the sight of 9/11 attack. Of course it is easy to cynically use 9/11 for political purposes. On the other hand, empathy is the very core moral value of the liberal worldview and the liberal parent’s loving response demonstrates the power of empathy to the hurt child, thus reinforcing the value further. Thus a Democrat is raised. Of course such a person will respond emotionally to the sight of Twin Towers falling. Of course such a person will find it appalling that such an event can be used as a pretext for starting an unwarranted war, using the event for political purposes, or even making the ad that was used in this study. Of course such a person would feel much greater and more urgent personal need to make sure that such an event never happens again, and will put much time and effort into learning how to make America safer and implementing the best possible strategy.
The inert, the moderate, the uninformed, the independents, whatever you may call them are quite capable of switching between the two modes. They may use one at home, another at work. What was their upbringing? A combo (e.g., Strict Mother and Nurturant Father)? Perhaps neglectful, or over-permissive childrearing? Single parent family? How do they respond to devastating events? How much empathy do they have? How will they respond to a possible October surprise? Who are they going to vote for on November 2nd? I wish I knew, and I wish the researchers have done a better job with their study so we could glean at least some ideas about these important questions.