Mixing Memory

Reading an article in the LA Times today, I learned something exciting: political differences in thought happen in the brain. At least that’s what a new study published in Nature Neuroscience(1) purports to show, though I hear that the next issue of the journal will contain critical responses from Descartes, Malenbranche, and Eccles.

Seriously though, the paper by Amodio et al. takes as its launching point the large body of evidence that political conservatives and liberals differ on personality dimensions related to openness to experience, tolerance of uncertainty, and cognitive complexity. It stands to reason that such differences in cognitive style manifest themselves in the brain (pace Descartes), but it doesn’t really suggest where in the brain. However, the relationship between these differences and conflict monitoring (a much more tenuous relationship, in terms of empirical support, I should note). Conflict monitoring is, in turn, seen as a form of cognitive control, or self-regulation, and is associated with a specific brain region: the anterior cingulate cortex (ACC; the blue part in the image below), part of the prefrontal cortex. Perhaps liberals and conservatives show differences in the activity of the ACC, then?

i-3c9f6c2f2b93b55da67670c6b971f067-ACC.JPG

Wait, what’s conflict monitoring? Some researchers have hypothesized that there’s a system in the brain tasked with looking for conflicts in information processing. This system is necessary, they argue, because with all of the processing going on inside our heads in parallel, there’s an inherent risk of “crosstalk interference” (2). That is, two streams of information that start out in different places might, at some point, cross paths. This leads to competition, conflict, and ultimately interference that can negatively affect performance. It’s important, then, to keep track of such conflicts, or potential conflicts at least, so that they can be avoided.

Now, this isn’t quite the definition of conflict monitoring that Amodio et al. put forward in their paper, but given that it’s the definition that the literature on conflict monitoring itself uses, and that the major theory linking cognitive control (and thus, self-regulation) to conflict monitoring uses this definition as well, I’m going to go with it. And to be honest, it’s not immediately clear how conflict monitoring in that sense and differences in liberal and conservative personality are related, but I can let that pass for now… OK, no I can’t really. Let’s face it, the intro to this paper, which is necessarily short because it’s only a “brief communication,” is terrible. I have no idea why they hypothesized a relationship between, say, tolerance for uncertainty and conflict monitoring. That hypothesis feels about as non sequitur as this:

Given that these associations between political orientation and cognitive styles [those described above - Chris] have been shown to be heritable, evident in early childhood, and relatively stable across the lifespan, we hypothesized that political orientation may be associated with individual differences in a basic neurocognitive mechanism involved broadly in self-regulation. (p. 1)

Well alrighty, then. Eye color is heritable, evident in early childhood, and relatively stable across the lifespan, but I’ve never seen someone associate it with self-regulation. Anyway, whatever the hidden logic behind the association of differences in cognitive style related to political orientation and conflict monitoring, here’s what they did.

They first had participants complete a political questionnaire, which included a question about whom, if anyone, they voted for in the 2004 presidential election, along with a scale ranging from -5 (“extremely liberal”) to 5 (“extremely conservative”). In previous research, that scale has correlated very strongly with voting behavior, and in their sample (43 right-handers), it was strongly correlated with 2004 voting. So it’s a pretty good measure of political orientation to the extent that liberals vote for Democrats and conservatives for Republicans.

Next they had participants complete a Go/No-Go task. In their version, participants were shown an M or a W, and told to respond only to one or the other by pressing a key (half the participants responded to M’s, and half to W’s). Responding to the target letter is the Go part, and not responding to the other letter is the No-Go part. In order to make responding the default, 80% of the 500 trials each participant saw were Go trials. This makes is so that when a No-Go trial pops up, participants face potential crosstalk interference. One system, the one that treats Go responses as the default, is trying to make the participant press the button, while the other system, recognizing that the No-Go letter has appeared, is trying to inhibit a response. As a result of this interference, the conflict monitoring system should kick in.

Now, we know that conflict in these sorts of tasks leads to increased activity in the ACC, so to look for differences between liberals and conservatives, Amodio et al. measured event related potentials (ERPs) there. And since they published the paper, you can probably guess that they found differences. Specifically, during the No-Go tasks, when the conflict monitoring system should be active and screaming, “Oh shit!” they found more neural activity in the ACC in liberals than in conservatives. Consistent with this, liberals were more accurate in the No-Go trials than conservatives. Thus they conclude:

Our results are consistent with the view that political orientation, in part, reflects individual differences in the functioning of a general mechanism related to cognitive control and self-regulation. Stronger conservatism (versus liberalism) was associated with less neurocognitive sensitivity to response conflicts. At the behavioral level, conservatives were also more likely to make errors of commission [i.e., pressing the button in No-Go trials -Chris]. (p.2)

Exactly what this means, I can’t really say. I’m not sure that it says much about the relationship between political orientation and self-regulation, but if it does, perhaps it provides a clue to the cognitive mechanisms underlying the personality variables associated with openness and uncertainty. The behavioral differences that result from those personality variables might also be related to this studies findings. Are conservatives more likely to make errors of commission in a wide variety of situations in which habitual responses are in place, for example? Only future research will tell. It shouldn’t, however, be taken as saying anything about the relative levels of intelligence of liberals and conservatives (though I’m pretty certain that it will, by people on both sides). So far, all it really says is that differences in thinking happen in the brain.

UPDATE: The Neurocritic gets all neurocritical.


1Amodio, D.M., Jost, J.T., Master, S.L., & Yee, C.M. (2007). Neurocognitive correlates of liberalism and conservatism. Nature Neuroscience.
2Botvinick, M.M., Braver, T.S., Barch, D.M., Carter, C.S., & Cohen, J.D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108(3), 624-652.

Comments

  1. #1 Joseph Hertzlinger
    September 10, 2007

    Have there been any studies on whether the Stroop effect is stronger for liberals or conservatives?

  2. #2 Hisham Zubi
    September 10, 2007

    Thanks for the overview.

    It would be interesting to see research that looks at how “conflict monitoring” might correlate to specific political issues beyond general political orientations.

  3. #3 Pelican's Point
    September 10, 2007

    I understand that the “conflict” discussed in the study was conflicts in internal information processing – not external conflicts that the person doing the processing is engaged in.

    However, that’s a possible processing difference that I’d like to understand better.

  4. #4 Anon
    September 11, 2007

    That’s strange. I thought that conservatism was positively associated with conscientiousness, conscientiousness was positively associated with self-regulation, and this was a self-regulation task.

  5. #5 Pelican's Point
    September 11, 2007

    This thread is pretty slow – so I’ll say more about what I was thinking to see if anyone else thinks about this stuff.

    Those are interesting findings IMO. They establish a possible neural basis for what is observed. I think however, that what we are discussing is behavior decisions – which is what brains are actually for.

    Specifically we are discussing the two human components of behavior decisions. One, thoughtful and calm, bringing in cognitive processes. The other, reactionary and instinctive (or defensive). That is – more reactionary and instinctive is how we tend to make behavior decisions in conflict. Conflict is where we feel a need to defend ourselves.

    Soldiers in combat do not expose themselves to enemy fire after calculating the cost benefit ratio of that action. They do it instinctively to protect their comrades. People in a crowd instinctively freeze and crouch down – when a bomb explodes nearby. They instinctively react to danger – no thinking involved.

    I think what the test shows is that conservative minds tend to act defensively as a default mode – even when there is little external conflict. It seems just participating in a test of some type is enough to make conservative minds enter defensive mode.

    And also perhaps, that liberal minds tend to not feel threatened so easily under similar relatively non-threatening stimulus – which probably allows liberal minds to bring more effective reasoning to bear on some decisions.

    I would guess though, that under conditions of more extreme conflict – that liberals can respond just as reactively as conservatives. We can shut off our brains too. I visit some “liberal” political forums – like DU – where knee jerk, reactionary (dumb) comments attacking other members can be very common.

    These reach their peak it seems prior to elections – when some members are being attacked for supporting “the wrong” Dem candidate by other members. i.e. their defenses are fully engaged for battle.

    It would be interesting to run the same study while imposing variable amounts of threat on the subjects – to see if and how the conservative and liberal subjects’ behavior decision modes change, respectively.

    It could even turn out that conservative minds – being more used to existing in the defensive state – could have a greater relative ability to apply reason under greater levels of threat.

    Or, maybe the most effective minds (behavior decision computers) are the minds that can switch between modes most appropriately for the type and level of threat that is sensed.

    I would note that cops and soldiers (who usually tend to be more conservative) are trained to remain calm and deliberative under very threatening situations.

  6. #6 doublehelix
    September 11, 2007

    Does anyone here question the findings of this study? Imagine how the subjects were recruited!? #1- the schools were UCLA and NYU, right? Think about the distribution of ideologies among the undergraduates at UCLA. Not exactly a hotbeds of conservatism/libertarian thought (although such schools are rare, I admit). #2-Imagine the posters used to recruit subjects. I wonder whether they were content neutral insofar as the aims of the study were concerned.

    “ARE THE BRAINS OF LIBERALS AND CONSERVATIVES DIFFERENT? EARN $50 TO FIND OUT!!!”

    Perhaps self-described liberals and conservatives were motivated by different things when they signed up for the study.

  7. #7 Garabaldi's Underwear
    September 11, 2007

    I ponder if there’s any difference between the animal spirits of Tories and Whigs. I’m flabbergasted that no man has set himself to discover it out!

  8. #8 John Broussard
    September 11, 2007

    It is useless to try to infer anything from this study. The experimenter has 26 liberals and 7 conservatives. Not exactly equal sample sizes, and not a big enough sample of conservatives to gain any inference about the range of responses here, even if you buy into the non-sequiter introduction set out by the authors. A thought experiment: would the correlation hold up if the two biggest outliers were excluded. Doesn’t look like it. No mention of possible sex differences etc. usually discussed in these papers. The authors fail to address these confounds even in the supplementary info. A sad commentary on the state of science when a journal with this level of impact publishes a study of this low quality.

  9. #9 cerebrocrat
    September 11, 2007

    This study, or more specifically, all the talk that’s going on about it, is an example of why I don’t like high-profile, general-discipline journals like Nature Neuroscience. There’s a lot of pressure, for career-promotion reasons, to get into these journals, and to do so, people have to sell their data like they’re soap flakes. A lot of the criticism of this study I’m seeing in the blogosphere treats it as if it’s supposed to be hypothetically critical, a make-or-break, close-the-book-on-this-subject, finding. It’s not, and scientific journal articles rarely are. It’s one more little piece in a huge puzzle, jazzed up a bit to help the lead author get a job. As a little piece in the puzzle, it’s a kind of interesting finding. There’s a fair amount of animal data on anterior cingulate function (btw, is AC really considered part of prefrontal cortex? I don’t think I’ve heard that before. Prefrontal cortex is phylogenetically newish, while I thought all that cingulate stuff was limbic, and thus much older in evolutionary terms.) and the real interest in this study, to me, is bridging the gap between what’s known from animal work and how it might apply to humans.

    As usual, (my bias shows through!) the animal data is explained in more concrete terms than “conflict monitoring.” (Try “discriminative stimulus learning”) Michael Gabriel at U. of Illinois has a career�s worth of animal data regarding anterior and posterior cingulate participation in stimulus discrimination learning. Briefly, if you give an animal two stimuli with the task of learning which one will predict a reward/punishment, the anterior cingulate seems to do the work of learning the predictive stimulus, while the posterior cingulate retains that stimulus for comparisons against others in future trials where the stimulus pairs are changed. These animal studies considered alongside this finding offer an interesting interpretation of conservatives� seeming commitment to fixed narratives: low sensitivity to newly predictive stimuli in favor of previously acquired associations. What�s more, the effect in animals is highly context-dependent: Using the same pair of stimuli, make one predictive in one context, but not another, and the anterior cingulate will discriminate between the exact same stimulus in the two different contexts � unless contextual information from the hippocampus is cut off via a lesion, etc.

    In the light of the animal data, the value of the Amadio study seems to be in how to take advantage of convergent evidence, and how to extrapolate from animal data to the much less constrained realm of human behavior outside the laboratory.

  10. #10 Neil
    September 12, 2007

    Scientific American :

    Are We Predisposed to Political Beliefs?
    Brain scans show that liberals and conservatives may be neurologically wired to lean politically left or right.

    Why don’t they just call it Unscientific American ?

  11. #11 Decline and Fall
    September 13, 2007

    I imagine that a lot of the problems in interpreting this data come from a poor understanding of what it means to be “liberal” or “conservative.”

    If what is being discussed is the general temperament associated with these terms (openness to new ideas being liberal, deference to tradition and authority being conservative) these results make intuitive sense.

    If the distinction is a contemporary political one these results don’t make much sense. The politically conservative position on taxation is that higher taxes are bad, no matter where you are. The temperamentally conservative position on taxation is whatever the dominant view of the day happens to be: a temperamentally consevative person in Sweden would find those agitating for lower taxes and fewer government services to be radicals.

  12. #12 Pat Mathews
    September 13, 2007

    Am I the only one reading this story who now can’t get Gilbert and Sullivan out of my head? “Is either a little Liberal – or a little Conservative!”

    Shuffles off this blog doing a sosft-shoe dance .

  13. #13 Lee
    September 17, 2007

    Fascinating article.

  14. #14 James Thivierge
    September 18, 2007

    Leave it up to the Liberals to come up with this kind of study and result! The Liberal mind is not only Biased, but manipulative and at best deceitful

  15. #15 Alex Leibowitz
    November 6, 2007

    If this is true, would it follow that individual political preferences are irrational?

  16. #16 Anon
    September 21, 2011

    The anterior cingulate would be picking up “conflicts” between information by seeing what small parts of it don’t fit together.

    If you tend to notice inconsistencies when you look at things more than the average person then you are more likely to notice problems in the way the world is, while conservatives are less likely to notice.

    I’d expect libertarians to line up with or more likely surpass the liberals rather than the conservatives. Most conservatives are not bothered by the incongruency between “small government” and “the war on drugs”. Libertarians are. Libertarians want small government and tolerate no inconsistencies, which suggests they have a very active anterior cingulate cortex.

  17. #17 JOSE DUNKER
    March 20, 2012

    I think the answer is “neuroplasticity”: the part of the brain that you use more, is going to develop more. In political activities with “liberal” polititians in my country (a caribbean island), I made the observation that this people share some basic traits: honest, responsible, conscious, and inflexible, all that related with ACC function, as the study sugest. In the other side, conservative could be more reactive, selfish, and aggresive, all traits related to amygdala function.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.