The claim: Politically liberal brains are better at handling change


A recent report in Nature Neuroscience has gotten a lot of press. The headlines proclaim that "left-wing" brains are different from "right wing" brains. Are our brains literally hard-wired to be conservative or liberal? The article in the L.A. Times sure seems to suggest it:

Sulloway said the results could explain why President Bush demonstrated a single-minded commitment to the Iraq war and why some people perceived Sen. John F. Kerry, the liberal Massachusetts Democrat who opposed Bush in the 2004 presidential race, as a "flip-flopper" for changing his mind about the conflict.

Really? Could one study of 43 college students actually tell us all that? Let's take a look at what the researchers, led by David Amodio, actually did find. Participants first took a survey in which one of the questions asked them to rate their political orientation on a scale ranging from "extremely liberal" to "extremely conservative."

Next they were attached to an electroencephalogram while they completed a quick go/no go task. The task is mind-numbingly simple, but it can be extremely difficult in practice. I've created a little movie (QuickTime required) so you can try it out. When you watch the movie, tap your finger on your desk every time you see the letter "W", and don't tap for any other letter.

Not as easy as it looks, is it? Now on to what the researchers found:

The W is shown 80 percent of the time, and the M is shown the rest of the time. After a while, it becomes quite difficult to resist tapping when you see the M. The researchers found a difference in accuracy between the viewers who said they were liberal and those who said they were conservative (viewers typed their responses on a keyboard instead of tapping on a desk). Liberalism correlated significantly (r = .30) with accuracy on the task. (Yes, the researchers controlled for the letters in the study -- liberals aren't only better at resisting when the letter is "W".)

They also found a difference in brain activity in two parts of the brain (actually this was measured in electromagnetic potential at the surface of the head, and the researchers extrapolated to find the location of the activity). Political orientation was correlated significantly with the strength of brain activity during the key no-go portion of the task (r = .41). The researchers calculated that the increased activity was occurring in the anterior cingulate cortex, responsible for conflict monitoring.

In other words, liberals are more likely than conservatives to have a strong response in the area of the brain used to inhibit responses at the time when they are supposed to inhibit response. So is this why Bush invaded Iraq and Kerry flip-flopped? Actually, Frank Sulloway, who made that claim, wasn't involved in the project. When the L.A. Times reporter finally interviewed Amodio, his response was much more guarded. According to the article, he "cautioned that the study looked at a narrow range of human behavior and that it would be a mistake to conclude that one political orientation was better."

The study results are actually quite modest -- the researchers claim to be the first to find a relationship between political differences and "a basic neuro-cognitive mechanism for self-regulation." However, these results are supported by a wide range of behavioral data which does support the idea that conservatives are less willing to accept complex arguments or shades of meaning compared to liberals.

Does this mean that liberals are "hard wired" to be different from conservatives? This data alone certainly doesn't support that claim. After all, the go/no go task is a learned activity. The reason that it's hard to inhibit tapping when the "M" appears is that you've learned to tap when you see a letter. You could also learn to tap only when you see an M, and you might be able to learn to be better at this task. Could you learn to be liberal (or conservative)? This study doesn't answer that question, but my suspicion is you could. After all, what's considered liberal in the U.S. is considered conservative in many places. In other places, a U.S. conservative would be considered a flaming liberal.

Amodio, D.M., Jost, J.T., Master, S.L., & Yee, C.M. (2007, Sept. 9). Neurocognitive correlates of liberalism and conservatism. Nature Neuroscience. DOI:10.1038/nn1979.


More like this

Maybe the release of this report is actually part of a covert study on how the critical reception of highly publicized, politically charged studies throughout the blogosphere is modulated by political preference.

I just have to say, that go/no go task is infuriating. I can't do it, and I'm the one who made it. I *know* when the M's are coming, and still I tap! Anyone else having this much difficulty with the task?

Did they control for any factors that may be related to training for this kind of response? Experience playing video games or sight reading music, for example, seem likely to have a large impact on how well a person would perform a task like this.

@MAttXIV: What reason is there to think that experience playing video games or sight reading music would correlate with political orientation?

Does this mean that liberals are "hard wired" to be different from conservatives? This data alone certainly doesn't support that claim.

someone fetch the twins...

btw, there have been several studies attempting to investigate the question of genetic contribution to behavioral modalities over the last few years.

there was a pretty well known twin study looking at the issue of extreme religious behavior using twin studies, for example.

I'm leaning towards the idea that there is in fact, a genetic contribution to the types of behaviors characterized as extreme political or religious viewpoints, or at susceptibility to same.

er, put "least" between "at" and "susceptibility".

btw, have you looked at the studies I was referring to, Dave?

I'd love to see a review if you did one.



Income and gender correlate with political views, so you have to factor those in as well. Gender is definitely correlated with video gaming, higher income school districts are more likely to have music programs, more politically liberal localities may fund music programs more, there seems to be a cultural norm of liberalism among musicians, etc. These correlations are a perenial problem with studying relations with political views.

Anyone who is seriously interested in this study owes it to him/herself to become familiar with the work of Michael Gabriel at U. of Illinois, Champaign-Urbana. Data from animals can be very useful for interpreting data from humans.

By cerebrocrat (not verified) on 11 Sep 2007 #permalink

Although I find the study interesting, I'm also deeply concerned about the connections we can draw between the data and the conclusions being drawn by some. Even if we were to establish that sex, socio-economic background, musical talent, gender, playing video games, etc were all standard, this study is still only studying college students (and only 43 of them at that). To draw any conclusions about the psychology of liberals and conservatives (without further qualification) we would need to conduct a study which studies liberals and conservatives without qualification. That is to say, we would need to study individuals from all walks of life.

Anyone who is seriously interested in this study owes it to him/herself to become familiar with the work of Michael Gabriel at U. of Illinois, Champaign-Urbana. Data from animals can be very useful for interpreting data from humans.

did you have a specific paper in mind?

I'm sure you couldn't be implying that liberals are better at/more accepting of mind-numbingly boring tasks like filling out government paperwork...

"...what's considered liberal in the U.S. is considered conservative in many places. In other places, a U.S. conservative would be considered a flaming liberal."

Also... we're relying on self-identification to separate the wheat from the chaff, err.. to separate the liberals from the conservatives. Independent measures of liberalism and conservativism would be necessary to underwrite conclusions about liberals and conservatives.

By bob koepp (not verified) on 12 Sep 2007 #permalink

"Yes, the researchers controlled for the letters in the study -- liberals aren't only better at resisting when the letter is "W"."

This is false. They "controlled" simply by reversing the letters. It was still a choice between "M" and "W". Also, I'm not sure what you're talking about in your concluding paragraph - "You could also learn to tap only when you see an M, and you might be able to learn to be better at this task." This is exactly what the study did.

In both instances, liberals are looking at the two letters with a different mindset than conservatives. And, as I pointed out here:
the study begins with a political questionnaire, predisposing the subjects to view "W" politically.

Perhaps testing Conservatives for basic literacy and/or the ability to draw conclusions from a simple graph or bar chart would be instructive.

What's interesting is how this relates to implicit association tests of this form:

The task in both bases is very similar, except that in the IAT you are selecting between two groups of categories (black and good vs. white and evil for example) instead of just between two categories.

However, it seems to me that if they have been primed by the questionnaire to consider W political, then a similar process is happening.

Also it's a shame that they chose W, since it's a common nickname for our current conservative president.

I think this also demonstrates, in the reaction to the study, that people seem to think that if its in the brain, it must be "hard-wired". Learned behaviors are also in the brain - and this study doesn't tell us whether the inhibition is hard-wired or soft-wired - it just tells us about the difference in behaviors and brain activity between the two groups.


I've voted (and held views) both conservative and liberal in my life. Earlier I worked alongside conservatives and found that I was (relatively) easily swayed to conservative groupthink (we didn't have too many alternative outlets available in the workplace).

However, later, with technology and and the commensurately expanding peer group, I tended toward the liberal. Also, others in that previously conservative workplace now tend toward the liberal as well, and they also attribute it to increased interaction with people and news sources that are available beyond the group bubble.

Go figure. But doesn't the fact that we change go counter to this notion of being hardwired?

I smell no smoke.


this is the popular press taking an oversimplification of the old ethological concept dating back to around the time of Konrad Lorenz and the early studies of animal instincts, and pasting on top of the idea of "genetic component".

people who don't know better (including the press) often confuse the issue of genetic contribution into thinking it means: preprogrammed and inflexible, when instead it usually refers to an underlying potential genetic component that might contribute to certain behavioral patterns being more or less likely to be adopted.

there are any number of behaviors that have been conclusively shown to have a significant genetic component, and still be flexible depending on various developmental/environmental circumstances.

think: schizophrenia, for example.

seriously, there is no reason to assume that behavioral components don't have a genetic component to them just like any morphological or physiological component does, and just like with a great many morphological components, environmental circumstances often have a great deal of influence as to how they are expressed.

Nature vs. Nuture is NOT the norm for the vast majority of traits studied to date; it's both nature AND nuture, to put it simply.

so, the press should entirely drop the usage of the word "Hardwired", since, by and large, it really doesn't explain what's going on in most cases, and even in the original usage, that of looking at repeatably observed instinctual animal behavior, there is still variability that can be induced.

It never ceases to amaze me that the press still choose to use words that are so archaic and misleading all the time.

Dave himself should be encouraging the dis-use of the term hard-wired, as you see even he puts scare quotes around it.

so stop it already, it just confuses an issue that was essentially resolved decades ago.

The liberal/conservative difference seems explainable by the construct of niche-picking (Scarr and McCartney, 1983), which holds that people actively choose environments that complement their genetic makeup. This explains why some developmental research has shown that identical twins reared apart can later be shown to have very similar hobbies, food preferences, and careers; also, why identical twins become more similar with age while fraternal and adopted twins become less similar. I'd bet in addition to the brain differences, there's some general personality difference between liberals and conservatives (however they were defined for this particular study).

It would be very interesting to see if there's a difference between the two political viewpoints using the Stroop effect paradigm--especially if the distractor words are ones typically associated with liberal or conservative viewpoints.

By Tony Jeremiah (not verified) on 12 Sep 2007 #permalink

"Does this mean that liberals are "hard wired" to be different from conservatives? This data alone certainly doesn't support that claim. After all, the go/no go task is a learned activity. ... Could you learn to be liberal (or conservative)? This study doesn't answer that question, but my suspicion is you could. After all, what's considered liberal in the U.S. is considered conservative in many places. In other places, a U.S. conservative would be considered a flaming liberal."

I agree that the study says nothing about whether the attitudes are learned -- but I suspect this particular comparison with other contries is way off the mark. Are we to believe that people from the Netherlands are be best at the task, Americans are somewhat worse, and Americans from 150 years ago would fail miserably? I doubt it. A more likely explanation is that the are two factors: (1) some individual underlying characteristic like willingness to accept nuances (either genetic or learned), and (2) the current local cultural zeitgeist. Political views stem from a combination of both factors, but performance in the test stems only or mainly from the first.

To summarize I'd expect that a person's liberality/conservativeness relative to their culture would be much better correlated with performance than a person's liberality/conservativeness measured on an absolute scale.

Something everyone is missing is that this study didn't measure a correlation in liberals and conservatives, but in people who identify themselves certain ways. What is a conservative? Does the modern college student who identifies with "conservative" really have much in common with either Pat Buchanan or Paul Wolfowitz in terms of personal intellectual history and philosophy? What this study is really showing is that people who don't deal well with change are people who identify themselves as not liking change. This study is worse than useless because it is so meaningless in its tautological quality that it's hard to believe it wasn't constructed that way on purpose.

And let me turn the interpretation on its head as well... We all know that academia is full of liberals from administration to professors to students: Two students coming from right-wing home backgrounds arrive as college freshmen, but one has, to quote the study abstract, "greater neurocognitive sensitivity to cues for altering a habitual response pattern." Which student is more likely to be responsive to cues as to how he is "supposed" to think? Maybe this study has just demonstrated the liberal disposition for group-think: the liberal ego is more malleable vis-a-vis the environment.

I found the task easy. Politically, I'm somewhere to the right of Genghis Khan.

BTW, Sulloway is hardly a disinterested observer. Four years ago, he coauthored a 'meta-analysis' with Jost, on of the authors of the present work, demonstrating that liberals are fluffy, cuddly, creative, charming, witty, intelligent, good looking, and scented like spring meadows; and conversely that conservatives are stupid, ignorant, misshappen, bloodthirsty, gratuitously cruel, evil, dungeon-dwelling mutants who trample on puppies for fun and recreation. Two of the four authors on that paper are contributors to Democratic candidates or leftist political organizations.

Of course, now you've got neurological effects demonstrated, you can come up with 'treatments' for conservatism, and put us away in nice friendly hospitals until we're 'cured'. It's not as if the left hasn't tried this before.

A little touchy on this subject, Gerard?

I agree, though, Sulloway's comment seemed quite biased to me. Having read the Nature Neuroscience study and supplemental materials, if Jost has any biases, they certainly don't come through in this study.

I do think the go/no go task might have very different results on different populations. For example, Greta tells me that there's some research suggesting that older adults are less able to inhibit responses.

Actually, no, I think it's rather funny; though the use of 'psychiatry' on political dissidents certainly wasn't. I was being sarcastic.

Mostly I think it's comical that someone has the chutzpah to pretend to use science to show that the people who agree with him are altogether superior people; and even more comical that some people (and I exempt you, Dave) apparently take it seriously.

Once again, Gerard, your own biases are showing.

do note the key response from Dave:

if Jost has any biases, they certainly don't come through in this study.

get it?

ready to stop being a psuedo-critic yet?

seriously, what exactly is the difference between you claiming this paper is based on nothing but bias, and an ID supporter claiming there is no ID "research" published because of bias?



get a clue, would ya? try to learn to control your own biases, or at least go do some research on how journal articles actually GET published, and how the peer review system works.

...or else be in danger of letting your own personal biases influence you so much that your commentary will be viewed as being worth exactly as much as the commentary from such luminaries as Salvador Cordova.

as i said to you on the PT thread, if you want to critique a paper and claim bias, you need to show how the methods used in the paper, and the analysis of the data itself is biased and incorrect, not just whine because you don't like the conclusions they or others draw from the data.

That's 'pseudo', fish-boy. The paper I was referring to was Jost, J.T., Glaser, J., Kruglanski, A.W., & Sulloway, F. (2003a). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-37, not to this study. The fact that Sulloway co-authored that paper, and not this one, would be a big clue. Reading comprehension is a valuable skill: I suggest you try to acquire it.

And if you insist on changing handles from forum to forum, don't blame me if I can't keep up. I have the cojones to post under my real name. Gerard there is Gerard here.

and, yet, it still is readily apparent to anyone with half a brain, Gerard, that regardless of which specific paper you are referring to, YOU STILL HAVEN'T done a blasted thing to actually look at the methods or results.

so, how can you possibly call yourself an unbiased reviewer, when it's obvious you haven't even examined, and likely don't have the background, to evaluate the methods used in the paper?

I'm going to keep calling you on this bullshit until you either:

actually read the papers, review the methods and results, and can prove bias and mistakes in method or results


you give up playing the armchair paper reviewer, and trying to use your background to lend weight to what amounts to nothing more than essentially scientific slander.

You should be frickin' ashamed to call yourself a scientist, and perhaps should be thinking about retirement if you think that your biases count as "analysis".

seriously, I see no difference between what you are doing here, and what the IDiots do when they claim their papers AREN'T published because of political bias.

you damn well know i have you dead to rights on this, so as the saying goes:

put up or shut up.


That's 'pseudo', fish-boy.

that's called a typo, pedant, and if you're trying to be insulting by pointing out typos on an internet forum, you're not only going to be labeled a pedant, but you're going to be awfully busy as well.

surely you have better approaches to criticism than pedantics?

...guess not.

Gerard must have decided on option 2.

It's sad when productive scientists get all old and fugly, and can't see when their analyses consist of paranoid projections of their own making.

I doubt, based on what's on Gerard's own site, that he is even capable at this point in time of self examination of his own extreme biases.

I do hope, if he reviews papers for journals himself (says he does), that he applies a more rigorous method than what he has shown here.

otherwise, he's in danger of providing the very fuel creationists use when they claim the reason none of their "work" gets published is because of the bias of the reviewers.

I would be curious to see how they defined liberal and conservative.

Socially liberal, economically liberal? Classical liberal as in libertarian? These words seem a bit too vague to have much useful meaning in an experiment, aside from telling us how people see themselves.

In my experience, there is immense social pressure in academia for students from conservative backgrounds to self report as liberal.

More importantly, if you were to create a second axis perpendicular to the left/right political spectrum, and called it Authoritarian/Libertarian spectrum, I think you might find some interesting results.

My guess, is people regardless of how liberal/conservative they are, who are more totalistic in thinking, would have more trouble distinguishing the Ms from the Ws.

There was an article on dogmatic thinking and the brain, that may have suggested something to that effect, but I can't recall it now. If anyone remember it feel free to post.

Students who play musical instruments must be better at quickly responding to a change in instruction or environment (eg. those who are good at sight-reading or impromptu accompaniments). Students who can play music may be more liberal.

Also, students who identify themselves as liberal tend to have more confidence in themselves. Confidence affects accuracy in performing a task.

Students who are highly sociable tend to be good at immediate adjustment according to other people's reactions. Students who have met a wide variety of people tend to identify themselves as liberal than those who only socialize with close friends who are similar to themselves.

But I also agree that it may be true that liberalism requires learning of quick adjustment. Conservative people tend to rely on tradition and do not like quickly changing environments. They may unconsciously train themselves to be more stable than flexible. I think liberal and conservative people have different strategies for survival.

By somereader (not verified) on 24 Sep 2007 #permalink