Spiders, MAggots, Politics

[Kanastrous]

Spiders, Maggots, Politics

A small but intriguing study finds that liberals and conservatives react differently when shown threatening images.

By Tony Dokoupil | Newsweek Web Exclusive

It's a golden rule of democracy that people are free thinkers. There's just one problem: we aren't wired that way—not, at least, according to a new study that probes past the rational mind in search of a biological basis for our political beliefs. The research, led by Rice University political scientist John Alford and published today in the journal Science, attempts to connect the dots between a person's sensitivity to threatening images—a large spider on someone's face, a bloodied person and maggot-filled wound—and the strength of their support for conservative or liberal policies. The subjects of the small but intriguing study were chosen through random phone calls to residents of Lincoln, Neb., and consisted of 46 mostly white Midwesterners who self-identified as having strong political beliefs. After filling out a survey of their political and demographic characteristics, participants were attached to a machine that measures arousal by increased moisture in the skin and presented with a slideshow of 30-odd images, including the three threatening ones. For comparison, the subjects were also shown three nonthreatening pictures—a bunny, a bowl of fruit and a happy child—within a separate sequence of slides. The more sensitive a participant was to the images, the wetter their skin got. Alford and his colleagues then correlated this with their political survey results.

The results seem to suggest that our ideas about the world are shaped by deep, involuntary reactions to the things we see. As evidence, the study found that greater sensitivity to the images was linked to more fervent support for a conservative agenda—including opposition to immigration, gun control, gay marriage, abortion rights and pacifism, and support for military spending, warrantless searches, the Iraq War, school prayer and the truth of the Bible. In other words, on the level of physiological reactions in the conservative mind, illegal immigrants may =s piders = gay marriages = maggot-filled wounds = abortion rights = bloodied faces. Before liberals start cheering, however, they don't come off much more noble or nuanced. They were less sensitive to the threatening images, and more likely to support open immigration policies, pacifism and gun control. But according to the research, that's hardly desirable, since it suggests that liberals may display mammal-on-a-hot-rock languor in the face of legitimate threats. "They actually don't show any difference in physical response between a picture of a spider on someone's face and a picture of a bunny," Alford tells NEWSWEEK. Alford spoke with Tony Dokoupil about the emerging connections between politics and the perceived intensity of threats. Excerpts:

NEWSWEEK: What surprised you the most?

John Alford: The clarity of the result. We came to this work after establishing that there is a genetic component to political ideology, so that made us interested in understanding how you get from the genes to political attitudes. One way into that territory was physiology. But I really didn't expect such crisp results.

Still, does the sample size worry you?

A small sample usually makes it harder to get to a level of statistical significance [because the results must be particularly one-sided to register]. So the fact that we [make it] is really quite powerful for a first attempt to explore this area. We'll follow up with larger groups, but what this needs more people for is to divide it out into different categories, rather than challenge the basic results.

On the level of physiological reactions, in the conservative mind, are spiders and illegal immigrants the same?

Physiology is a blunt way to go from attitudes into biology, so it's not clear exactly how those two things might connect. But that's possible. The immediate way we experience threats might predispose people to find socially protective policies [like tight border control] more or less persuasive. For people that have low levels of support for these socially protective policies, on the other hand, they actually don't show any difference in physical response between a picture of a spider on someone's face and a picture of a bunny. So a way of thinking about this is that there's a portion of the population that is physically primed to be alert to threat, and that feels it physically, and they're more likely to support a set of positions shaped by that response to threat.

Other studies have focused on the psychology, rather than biology, of political beliefs. How are the two strains of research related?

All combined, they show us that there are roughly three influences on political opinion. One is a biological predisposition. Our study is a small window into that. Another is traditional socialization, such as the fact that I grew up during the depression, was in a lower middle-class family, and my parents were Republicans. The last is adult experience, reasoning power, or what's traditionally called free will.

How do those work out in practice?

If you ask someone why they support the Iraq War, they would probably give you some answers out of those latter two categories. They would make an intellectual argument: we were faced with a threat and this was the right choice. If you pushed, they might also mention socialization: well, I'm an Army brat, my dad was a colonel, my brother's in the Marines. One thing that they'd never say, in my experience is I'm simply biologically predisposed to be sensitive to threats. What's really important here is that we're not dismissing intellectual choice or experience. We're just asking for a place at the table for biology.

If political beliefs are hardened by biology, how do you explain flip-flops?

You'd only have trouble explaining them if you considered biology to be deterministic. By the same token, if you thought childhood socialization was everything, you couldn't explain a flip-flop. The person's childhood didn't change.

If biology isn't deterministic, is it at least probabilistic?

It's a question of how easy it is to get from event A to belief B. Russia invades Georgia. What's your response? Military action or be nice to the Russians? You can get to either position intellectually, but how easy it is influenced by whether you experience it as an immediate physical threat or not. I don't think that biology is destiny, but for the general public, I want people to believe that it's something. Right now it's seen as nothing. It's given zero weight.

Tell that to the people making political ads with packs of wolves and ominous 3 a.m. phone calls.

That's interesting, actually, because this study shows that the ability for [scare tactics] to work may not be uniform across the population. As for why political strategists have long used threatening images, it seems that sometimes people who do something have a sort of folk wisdom that exceeds the general knowledge, and even the academic knowledge.
You often hear that the right is great at "mobilizing their base."

Could this be because the right is more sensitive to threats?

I think that's one conclusion. It may also explain why it's self-apparent to people who hold [what are now right-wing positions] that they're really important, and frustrating why it isn't obvious to the other side. It's like, "What part of the difference between a spider and a bunny don't you understand."

[Akhlut]

Interesting, but I think they could have done the experiment better.

Firstly, some people might not find spiders, maggots, or people with bloody faces threatening. For instance, I'm actually fond of spiders, and while I find maggots distasteful, it's not a huge reaction from me, unless they were on something I had direct interest in that I didn't want to study.

Secondly, while pictures can be handy for gauging reactions, people living in modern societies are well-acquainted with pictures, and unless they simply flashed the image on the screen for a second or two, it might be likely that people who saw threatening images quickly understood them to be just pictures, and their physiological mechanisms didn't engage as much.

Thus, while the study seems to have a firm enough base, I'd like to see the experiment ran again with some differences, like presenting people with a number of objects and perhaps animals in cages. Perhaps a things like a flower, skillet, light bulb, headphones, crackers, rabbit, snake, and a bloodied knife.

[Alyrium Denryle]

Interesting, but I think they could have done the experiment better.

Firstly, some people might not find spiders, maggots, or people with bloody faces threatening. For instance, I'm actually fond of spiders, and while I find maggots distasteful, it's not a huge reaction from me, unless they were on something I had direct interest in that I didn't want to study.

Secondly, while pictures can be handy for gauging reactions, people living in modern societies are well-acquainted with pictures, and unless they simply flashed the image on the screen for a second or two, it might be likely that people who saw threatening images quickly understood them to be just pictures, and their physiological mechanisms didn't engage as much.

Thus, while the study seems to have a firm enough base, I'd like to see the experiment ran again with some differences, like presenting people with a number of objects and perhaps animals in cages. Perhaps a things like a flower, skillet, light bulb, headphones, crackers, rabbit, snake, and a bloodied knife.

That is why it is harder, with small sample sizes, to get statistically significant results. They got significant results through all those issues, when you otherwise would not expect to. The magnitude of the differences are huge and you would expect things like being fond of spiders to affect either group the same way

[Akhlut]

That is why it is harder, with small sample sizes, to get statistically significant results. They got significant results through all those issues, when you otherwise would not expect to. The magnitude of the differences are huge and you would expect things like being fond of spiders to affect either group the same way

True; another problem is the small and similar sample size of the study: 46 white Nebraskans. While they might have divergent political opinions, I imagine their life experiences to be fairly similar. Hell, the picture of the bloodied face may have induced a reaction based on the race of the person pictured, rather than the fact that the person had blood on their face.

[Patrick Degan]

I heard about this study on NPR Friday morning. It seems to me that it started out attempting to prove a pre-determined conclusion about human behaviour and politics and proceeded upon some rather flawed methodology. There is no baseline set of reactions to stimuli to act as a comparison, no control group, and a failure to take into account that startle-reactions are as much individual as any other aspect of human psychological and intellectual makeup. So a person gets startled when somebody suddenly pumps a burst of loud static through the headphones the moment a picture of a gun flashes up on screen and that proves what, exactly? How is disgust at a picture of maggots or a spider on somebody's face supposed to indicate political leanings, exactly? You bombard a person with negative stimuli to get a startle/fear/disgust reaction long enough and slip in a picture of a bunny or an abortion clinic or two gay people kissing when the subject is still reacting to the other images and what exactly is that supposed to demonstrate?

Did Alford and his researchers include any sort of downtime/recovery period for any given stimulus as part of his reaction tracking? Or did he just lump it all together through a session with any given subject and average out the results? (addendum)And then you've got the problem of having a very small sample group from a rural state, which cannot yield anything like an accurate picture of a large population across the ideological spectrum.

I'm sorry, but this experiment sounds like bullshit.

[Skgoa]

[Liberals] were less sensitive to the threatening images, and more likely to support open immigration policies, pacifism and gun control. But according to the research, that's hardly desirable, since it suggests that liberals may display mammal-on-a-hot-rock languor in the face of legitimate threats.

bullshit. this characteristic is EXTREMELY desirable. also, it kind of shows, what this "researcher's" own political oppinios are.

"They actually don't show any difference in physical response between a picture of a spider on someone's face and a picture of a bunny," Alford tells NEWSWEEK.

that is because a spider on someone else IS NOT THREATENING. the best reaction to it is STAYING absolutely CALM, telling the other one to stand still and removing the spider. its the same with maggots in a wound: they are no emediate threat (they might even help clean the wound) and can be removed without danger to anyone. and blood an someone's face: a) he is injured, in wich case I have to STAY CALM and help him, b) its someone else's blood, in which case I have to STAY CALM and help that guy or c) its the blood of an animal or fake blood or anything else that does not require an emediate reaction from me... which means I should STAY CALM. noone with at least two braincells should feel threatened by any of that.

so, what does the study proof:
liberals react in a logical way, even on the most basic emotional level. the conservatives on the other hand panic easily. now I know who I want to pick up the phone at 3 a.m....

[Terralthra]

I heard about this study on NPR Friday morning. It seems to me that it started out attempting to prove a pre-determined conclusion about human behaviour and politics and proceeded upon some rather flawed methodology. There is no baseline set of reactions to stimuli to act as a comparison, no control group, and a failure to take into account that startle-reactions are as much individual as any other aspect of human psychological and intellectual makeup. So a person gets startled when somebody suddenly pumps a burst of loud static through the headphones the moment a picture of a gun flashes up on screen and that proves what, exactly? How is disgust at a picture of maggots or a spider on somebody's face supposed to indicate political leanings, exactly? You bombard a person with negative stimuli to get a startle/fear/disgust reaction long enough and slip in a picture of a bunny or an abortion clinic or two gay people kissing when the subject is still reacting to the other images and what exactly is that supposed to demonstrate?

Did Alford and his researchers include any sort of downtime/recovery period for any given stimulus as part of his reaction tracking? Or did he just lump it all together through a session with any given subject and average out the results? (addendum)And then you've got the problem of having a very small sample group from a rural state, which cannot yield anything like an accurate picture of a large population across the ideological spectrum.

I'm sorry, but this experiment sounds like bullshit.

I think you should have actually read the article. The opinions on gay rights, abortion, etc., were from a survey that each participant filled out. The correlation is that those who responded strongly to visually 'scary' pictures like giant spiders, maggots, etc., all had strong politically conservative leanings in the opinion survey they filled out, while those who leaned strongly liberal did not have the same set of strong reactions to those images.

[Alyrium Denryle]

That is why it is harder, with small sample sizes, to get statistically significant results. They got significant results through all those issues, when you otherwise would not expect to. The magnitude of the differences are huge and you would expect things like being fond of spiders to affect either group the same way

True; another problem is the small and similar sample size of the study: 46 white Nebraskans. While they might have divergent political opinions, I imagine their life experiences to be fairly similar. Hell, the picture of the bloodied face may have induced a reaction based on the race of the person pictured, rather than the fact that the person had blood on their face.

Except that when you use that particular methodology, you are also controlling for every variable but what you are looking at. Political leanings.

For a sample size that small, you have to do that to have any resolution. You can't generalize, and they admit this. But it gives you the preliminary data needed to get better grant funding

[xerex]

so this study proves conservatives get scared by pictures (ie non real threats) while liberals stay clam ?

[Kanastrous]

I think it's more that threatening stimuli (the pictures) is more threatening to conservatives, than to liberals. Within the sample group studied.

[Patrick Degan]

I heard about this study on NPR Friday morning. It seems to me that it started out attempting to prove a pre-determined conclusion about human behaviour and politics and proceeded upon some rather flawed methodology. There is no baseline set of reactions to stimuli to act as a comparison, no control group, and a failure to take into account that startle-reactions are as much individual as any other aspect of human psychological and intellectual makeup. So a person gets startled when somebody suddenly pumps a burst of loud static through the headphones the moment a picture of a gun flashes up on screen and that proves what, exactly? How is disgust at a picture of maggots or a spider on somebody's face supposed to indicate political leanings, exactly? You bombard a person with negative stimuli to get a startle/fear/disgust reaction long enough and slip in a picture of a bunny or an abortion clinic or two gay people kissing when the subject is still reacting to the other images and what exactly is that supposed to demonstrate?

Did Alford and his researchers include any sort of downtime/recovery period for any given stimulus as part of his reaction tracking? Or did he just lump it all together through a session with any given subject and average out the results? (addendum)And then you've got the problem of having a very small sample group from a rural state, which cannot yield anything like an accurate picture of a large population across the ideological spectrum.

I'm sorry, but this experiment sounds like bullshit.

I think you should have actually read the article. The opinions on gay rights, abortion, etc., were from a survey that each participant filled out. The correlation is that those who responded strongly to visually 'scary' pictures like giant spiders, maggots, etc., all had strong politically conservative leanings in the opinion survey they filled out, while those who leaned strongly liberal did not have the same set of strong reactions to those images.

You're still dealing with a very small sampling group and one apparently polarised either left or right in their beliefs, though, and that's nowhere near enough to establish a definitive analysis. I did misunderstand the details of the survey, but even reading the article in greater depth this still points to some questionable methodology on Alford's part.

Alford is hanging his hat on the idea that conservatives "are primed to be more sensitive to threats" than liberals. But the observation that conservatives see threats more that liberals do leans far more towards psychology than biology. Threaten a liberal in his face, in the real world, and he'll perceive the threat as immediately and have the same fight-or-flight response as the conservative will. But it's a very long leap to suggest that a conservative tendency to see threats in a lot more places or feel them more intensely than liberals will indicates a greater biological priming in conservatives to threat perception as opposed to the "mammal-on-a-hot-rock langour" by liberals who don't get the sweats from a picture of a spider.

[Akhlut]

Except that when you use that particular methodology, you are also controlling for every variable but what you are looking at. Political leanings.

For a sample size that small, you have to do that to have any resolution. You can't generalize, and they admit this. But it gives you the preliminary data needed to get better grant funding

Shit! How could I forget that? I take four years' worth of biology and related classes and I forget the need to control for variables; how the fuck did I forget that basic tenet of science?


As for sample size: couldn't they have gotten a few dozen more volunteers, though, just to reach about 100 white Nebraskans? It might just be me, but I like chi squares (or whatever statistical analysis you want to use; but I think with a simple test like this that has one variable, a chi square would be adequate, though a T-test would also be effective, but it's been a couple years since I took statistics, so I admit to the possibility of overlooking something more effective) that are more definitive.

[xerex]

I think it's more that threatening stimuli (the pictures) is more threatening to conservatives, than to liberals. Within the sample group studied.

except its still a picture.

as said above they really should have used live threats. they havent controlled for the possiblity that some recognize that a picture no matter how gruesome is no threat.

[Alyrium Denryle]

I think it's more that threatening stimuli (the pictures) is more threatening to conservatives, than to liberals. Within the sample group studied.

except its still a picture.

as said above they really should have used live threats. they havent controlled for the possiblity that some recognize that a picture no matter how gruesome is no threat.

You guys are bitching about a non-issue, because that problem would have equally confounded the results in both treatment groups.

And by your logic, every study that uses photos instead of live action stuff including most uses of brain imaging techniques are invalid. Which is not the case, nor would porn have the physiological effect that it does.

You might know it is a picture, but your basic physiological responses to stimuli do not.

but it's been a couple years since I took statistics, so I admit to the possibility of overlooking something more effective) that are more definitive.

They probably used an ANOVA variant

[Akhlut]

You guys are bitching about a non-issue, because that problem would have equally confounded the results in both treatment groups.

And by your logic, every study that uses photos instead of live action stuff including most uses of brain imaging techniques are invalid. Which is not the case, nor would porn have the physiological effect that it does.

You might know it is a picture, but your basic physiological responses to stimuli do not.

I don't doubt the efficacy of photos that aren't shown for long periods of time, as the brain registers them essentially the same as a real object; however, the longer it is shown, the more issue I'd take with it, due to the fact that a photo is static. So, if the study only flashed an image for 5 seconds or less (which it almost certainly did), I'd not have a problem with it. Although, I'd have to discuss the issue with my brother (who has a master's degree in neuroscience and worked on facial recognition software, so, I assume he might know how static images are perceived in the brain).

I think I'm just more interested in how the groups react when presented with a "longer term" danger, which I think would be best reflected by a real object, not a picture, and going off the physiological responses from that, in addition to the pictures (because, hey, I'm all about getting more data).

Of course, if the study had little funding or if a lot of it had to go to "administrative overhead" *cough*, I can understand them not going to the lengths and expense of getting live tarantulas, rabbits, etc.

but it's been a couple years since I took statistics, so I admit to the possibility of overlooking something more effective) that are more definitive.

They probably used an ANOVA variant

Sadly, I took only a single semester of stats, so we only went over the basics (no ANOVA). Thus, I'm assuming I'll get to learn all about ANOVA when I can get into grad school and have to remedy my math deficiencies.

[Covenant]

It's an interesting study but it's really flawed to have such a small size--I've got family over in Lincoln, and if they were chosen, I can tell you several reasons why seeing a spider might not make them freak out. It's possible that the big difference between Liberals and Conservatives is horror movie watching and 4chan. In any case, there needs to be a much wider sample taken after this, since the stimuli in question is so unequally disturbing. I find it hard to imagine I would react negatively to a picture of a spider--especially when I use them for reference when making art, but I find spiders absolutely fucking horrifying in real life.

If you want to test people's reactions to threatening stimuli, there's other ways to do it that test for the stimuli rather than the medium.

[Alyrium Denryle]

It's an interesting study but it's really flawed to have such a small size--I've got family over in Lincoln, and if they were chosen, I can tell you several reasons why seeing a spider might not make them freak out. It's possible that the big difference between Liberals and Conservatives is horror movie watching and 4chan. In any case, there needs to be a much wider sample taken after this, since the stimuli in question is so unequally disturbing. I find it hard to imagine I would react negatively to a picture of a spider--especially when I use them for reference when making art, but I find spiders absolutely fucking horrifying in real life.

If you want to test people's reactions to threatening stimuli, there's other ways to do it that test for the stimuli rather than the medium.

How many times must I repeat myself... the magnitude of the difference between groups has to be HUGE to get a statistically significant difference with this small a sample size. The fact that it is there anyway, means that the small sample size is irrelevant, and chance factors, like the random variation in perusal of 4chan, or someone liking spiders in the sample groups, are not the cause for the difference seen, especially because confounding factors like that are equal on both sides.

Do I have to put that in huge bold letters? Am I the only one in this thread to take a stats class or ever run a statistical test on a real data set?

[Covenant]

We'll see. His twin studies were interesting, but I haven't seen the Data yet. Does anyone get Science? If this is what the research actually shows then I'll agree, but it just doesn't smell right without some numbers and context. That's a subjective criticism, but I find it hard to accept, especially with so many other variables--such as income, age, gender, and occupation--giving so many possible confounding factors a say in such a small sample. And I think it's fair to say that politics are not equal amongst all those variables, especially in Lincoln, Nebraska. This isn't to say you can't establish a decent sample size that's small, but this seems too uncontrolled for something as particular as politics and I still haven't seen the actual results in anything other than interview form.

You don't think it's reasonable to cast some doubt on the veracity of the research?