follow CCP

Recent blog entries
popular papers

Science Curiosity and Political Information Processing

What Is the "Science of Science Communication"?

Climate-Science Communication and the Measurement Problem

Ideology, Motivated Cognition, and Cognitive Reflection: An Experimental Study

'Ideology' or 'Situation Sense'? An Experimental Investigation of Motivated Reasoning and Professional Judgment

A Risky Science Communication Environment for Vaccines

Motivated Numeracy and Enlightened Self-Government

Making Climate Science Communication Evidence-based—All the Way Down 

Neutral Principles, Motivated Cognition, and Some Problems for Constitutional Law 

Cultural Cognition of Scientific Consensus
 

The Tragedy of the Risk-Perception Commons: Science Literacy and Climate Change

"They Saw a Protest": Cognitive Illiberalism and the Speech-Conduct Distinction 

Geoengineering and the Science Communication Environment: a Cross-Cultural Experiment

Fixing the Communications Failure

Why We Are Poles Apart on Climate Change

The Cognitively Illiberal State 

Who Fears the HPV Vaccine, Who Doesn't, and Why? An Experimental Study

Cultural Cognition of the Risks and Benefits of Nanotechnology

Whose Eyes Are You Going to Believe? An Empirical Examination of Scott v. Harris

Cultural Cognition and Public Policy

Culture, Cognition, and Consent: Who Perceives What, and Why, in "Acquaintance Rape" Cases

Culture and Identity-Protective Cognition: Explaining the White Male Effect

Fear of Democracy: A Cultural Evaluation of Sunstein on Risk

Cultural Cognition as a Conception of the Cultural Theory of Risk

« "Gateway belief" illusion--published and critiqued (?) | Main | Hey, want to know someting? Science curiosity is a culturally random variable! »
Saturday
Dec092017

A draw in the “asymmetry thesis meta-analysis” steel-cage match? Nope. It’s a KO.

As the 14 billion regular subscribers to this blog know all too well, I’ve been discussing the so-called “asymmetry thesis” (AT) on this site (and in published papers [Kahan 2013]) for approximately 65 years now.

AT posits that the impact of ideologically motivated reasoning is asymmetric in relation to so-called “liberal” and “conservative” orientations. Conservatives, AT proponents maintain, are substantially more vulnerable to this form of biased information processing than are liberals (e.g., Jost et al 2003).

What about AT opponents? What do they say?

Well, I don’t recall any empirical researcher who asserts that liberals are more biased than conservatives (maybe motivated reasoning is causing me to overlook or just not recall such research).

Rather, AT opponents contend politically motivated reasoning is uniform—i.e., symmetric—across the conventional left-right spectrum.  So let’s call this position “ST” for “symmetry thesis.”

The fight between AT and ST looks like the kind of dispute that ought to be adjudicated by meta-analysis.  And in fact, in the last 6 mos. or so, we’ve been treated to two meta-analytic investigations, one by John Jost (2017) and another by Pete Ditto & a large contingent of collaborators (in press).

The problem, however, is that Jost and Ditto et al. appear to strongly disagree with one another about what their massive literature surveys imply.

Jost reports finding approximately 280 studies involving almost 400,000 subjects. From the “need for closure” to “dogmatism” to “self-deception”—the self-report measures featured in these studies support the conclusion that conservatives are more biased than are liberals.

Meanwhile, Ditto et al. report the results from 51 experiments, comprising 18,000 subjects. Their conclusion? That “there was strong support for the symmetry hypothesis: liberals (r = .235) and conservatives (r = .255) showed no differnce in mean levels of bias across studies"—a compelling affirmation of ST over AT.***

So now what? Do we just throw up our hands and give up?

The answer is no. It turns out that Jost’s and Ditto et al.’s results can be reconciled pretty easily. All one has to do is examine what they were measuring and how.

Jost’s meta-analysis was based on survey data correlating conservatism and various measures of cognitive style.  Jost did not present any meta-analytic data on motivated-reasoning experiment results.

That’s what Ditto et al. measured.  They included in their sample, moreover, only experimental studies that conformed to the Politically Motivated Reasoning Paradigm (“PMRP”). PMRP identifies a method specifically crafted to avoid the myriad confounds that can rob a study of politically motivated reasoning of its validity (Flynn et al. 2017; Johnston & Ballard 2016; Kahan 2016a).  Focusing on studies that meet the PMRP standard, Ditto et al. conclude that liberals and conservatives were equally vulnerable to politically motivated reasoning.

More or less as an aside, Jost does refer to several experimental studies in his paper. But he doesn’t say anything about the criteria he used for singling them out, much less about whether they were consistent with PMRP.

Indeed, it’s clear that the main criterion Jost used to flag these particular experimental studies was that they reached a result congenial to his hypothesis.  We can tell that he resorted to cherry-picking of this sort * because he didn’t cite a single one of the myriad experimental studies that suggest that liberals are as prone to ideologically motivated cognition as conservatives.  We know there are many studies like that because plenty of them were featured in Ditto et al., an earlier version of which is in fact cited by Jost.****

There’s no reason, though, to doubt that Jost used appropriate criteria, applied with appropriate impartiality and care, to select studies that report the relationship between liberal-conservative ideology and one or another self-report measure of cognitive style.

But that only makes things worse for AT.  For notwithstanding the preponderance of evidence that conservatism is associated with a closed-minded style based on “epistemic” self-report  measures, Ditto et al. demonstrate that liberals are every bit as likely to succumb to politically motivated reasoning when one tests partisans’ information processing experimentally. This combination of results, then, implies that the self-report measures Jost analyzes are externally invalid indicators of what we actually care about—viz., how individuals of opposing political outlooks actually process information.

The only objective reasoning-style disposition that Jost reports on is the Cognitive Reflection Test (CRT), on which liberals, according to Jost, have a modest performance advantage over conservatives.

But here, too, Jost’s fixation on correlational studies and his resolute disregard for experimental ones undermines his conclusions. MS2R—“motivated system 2 reasoning”—describes the tendency of those who score highest on objective measures of cognitive proficiency (including not only CRT but also Numeracy and Ordinary Science Intelligence) to display more bias, not less, when they process political information (Kahan 2016b).

Thus, if we take Jost’s compilation of studies featuring CRT at face value, his finding that liberals score higher on it is a reason to infer that liberals are more vulnerable, not less, to politically motivated reasoning than are conservatives.

But we shouldn’t do this.

If one is trying to figure out who is more disposed to process political information in a biased manner-- conservatives or liberals—one should examine how they actually reason.

Ditto et al. do this.  Jost doesn’t.

Thus, the “meta-analysis steel-cage match” was no tie. 

On the contrary, it was a knock-out victory for ST over AT.

Refs

Ditto, Peter H. and Liu, Brittany and Clark, Cory J. and Wojcik, Sean P. and Chen, Eric E. and Grady, Rebecca Hofstein and Zinger, Joanne F., (in press). At Least Bias Is Bipartisan: A Meta-Analytic Comparison of Partisan Bias in Liberals and Conservatives. Perspectives on Psychological Sci. Working paper available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2952510.

Flynn, D. J., Nyhan, B., & Reifler, J. (2017). The Nature and Origins of Misperceptions: Understanding False and Unsupported Beliefs About Politics. Political Psychology, 38, 127-150. doi: 10.1111/pops.12394

Johnston, C. D. and A. O. Ballard (2016). "Economists and Public Opinion: Expert Consensus and Economic Policy Judgments." The Journal of Politics 78(2): 443-456.

Johnston, C. D., & Ballard, A. O. (2016). Economists and Public Opinion: Expert Consensus and Economic Policy Judgments. The Journal of Politics, 78(2), 443-456. doi: 10.1086/684629

Jost, J. T. (2017). Ideological Asymmetries and the Essence of Political Psychology. Political Psychology, 38(2), 167-208.

Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political Conservatism as Motivated Social Cognition. Psych. Bull., 129(3), 339-375.

Kahan, D. M. (2013). "Ideology, Motivated Reasoning, and Cognitive Reflection." Judgment and Decision Making 8: 407-424.

Kahan, D. M. (2016a). The politically motivated reasoning paradigm, part 1: What politically motivated reasoning is and how to measure it. Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource.

Kahan, D. M. (2016b). The politically motivated reasoning paradigm, part 2: Open questions. Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource.

* John convinced me that the stricken language comes across as asserting that he engaged in wrongdoing, which is not what I meant to assert.  My point is that he cites the experiments in question for illustration, not for proof that experimental studies show the asymmetry that he reports for cognitive-disposition measures.

** Not in original post.

*** Revised to reflect "in press" version of Ditto et al.

**** John still (reasonably) objects to the discussion of his treatment of experiments in the paper. I included that discussion only b/c I anticipated John would point out that he did look at experimental evidence too (albeit by non meta-analytic techniques). But the post doesn't require the relevant paragraphs  to make its points--none of which is to imply that John acted in bad faith.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (47)

And a knock out victory for objectivity.

I've not delved into this topic, but I've wondered on occasion what on Earth causal reason AT supporters might propose that could support their position. Granted they don't need this first, but they would need such an underpinning eventually. Cultural diversity both now, and even more so in history and pre-history, is vast. While to contemporary Western eyes current US conservative and liberal cultural modes seem very different, they have much more in common with each other than they do with most of these cultures, the vast majority of which are lost to history yet of which we have surviving evidence fragments pointing to very wide diversity indeed. The US cons and libs are just two arbitrary emergent points on an enormous cultural topography, and relatively local at that. Yet for one to systemically invoke much more motivated reasoning than the other, implies deep mechanisms that would at the very least have to come out of long cultural evolutionary timescales, if not still longer gene / culture co-evolution timescales (which would also imply that libs and cons are actually different types of people - the conservative brain etc). But this swims against the entire tide of evolutionary literature. The underlying mechanisms of all cultures and our pre-sensitization to them, are the same; hence it is very hard to see how significantly different biases or levels of bias could occur in folks who are equivalently cultural but within different cultures, whether con / lib or any other.

December 9, 2017 | Unregistered CommenterAndy West

I'm not a proponent of the asymmetry theory, but...

If one is trying to figure out who is more disposed to process political information in a biased manner-- conservatives or liberals—one should examine how they actually reason.

This, of course, assumes that the inclusion of experimental data is dispositive. I don't doubt that it is generally preferable in many ways, but a blanket assumption that inclusion of such = a KO seems quite dubious to me. Much depends on the quality of the experimental paradigms, and whether or not they have external validity, are truly valid w/r/t generalizing to the real world. It certainly isn't hard to find many, many experimental studies which might have strong internal validity but questionable external validity.

December 9, 2017 | Unregistered CommenterJoshua

"I've not delved into this topic, but I've wondered on occasion what on Earth causal reason AT supporters might propose that could support their position."

There are lots of possibilities. Reasoning characteristics can correlate with political orientation either by people with a mental characteristic being attracted to a particular sort of politics, people with a certain political orientation being attracted to or more exposed to certain reasoning styles, environmental factors like education or parental upbringing passing on both reasoning styles and political assumptions, cultural transmission of both reasoning and politics simultaneously, and no doubt many others.

Some of the mental characteristics that have been or might be proposed include:

Openness to new experiences/knowledge versus respect for tradition. Open people are inclined to like things just because they're new and unfamiliar.

'Hierarchical' respect for authority versus 'individualist' liking for the maverick outsider. Some people automatically trust big institutions, impressive credentials, age and experience. They assume the standard way of doing it is most likely correct. Others believe in intellectual revolutions, new discoveries overturning old dogmas, and continual change, and tend to automatically distrust Establishments and authority figures.

'Empathic' reasoning from consequences versus 'cold' logic. Some people will shy away from a line of reasoning if it leads to unpleasant or immoral consequences for people they sympathise with. (It's often based on the underlying "just world hypothesis" that correct reasoning must give rise to a morally acceptable result, therefore immoral consequences mean the reasoning must somehow be incorrect, even if you can't immediately see the flaw.) Others are more willing to let the logic take them where it will, even if the conclusions are morally repugnant (which can of course lead them to miss reasoning errors that a closer examination would have found).

System 1 and System 2 processing. Some people always go with their initial 'gut' impression, others are more inclined to to process questions analytically and check the results.

Analytic versus synthetic reasoning. Some people prefer to take a complex effect and break it down into its basic components. Other people prefer to take simple and easily understood effects and combine them into a more complex model, elaborating gradually.

Simple single-step versus complex multi-step causality. Some people like to boil everything down to a single simple explanation/mechanism, with a simple and obvious answer/policy, and will ignore any complexities or additional consequences as irrelevant distractions. Other people are sceptical of simplicity, and always look for extended consequences and competing effects.

Concrete versus abstract reasoning. Some people can follow reasoning with specific, familiar examples and cases, but cannot extend the same principles to analogous cases and situations.

Trusting friends versus trusting outside experts. Some people base their trust on familiarity and long experience with people they know well, and distrust strangers. Others vice versa.

But I think most people who hold the belief don't reason that far. The basic argument goes: "There are some policy positions that are factually and morally *obvious*, to the point where it's inconceivable that any rational person could disagree with them. But those cultural outsiders over there don't believe in them. Therefore those people cannot be rational, and it's just a matter now of figuring out what specifically is wrong with their reasoning ability." The fact they haven't done so yet doesn't invalidate their basic argument.

"What about AT opponents? What do they say?

Well, I don’t recall any empirical researcher who asserts that liberals are more biased than conservatives (maybe motivated reasoning is causing me to overlook or just not recall such research)."

Such a person wouldn't be an AT opponent - they'd be an AT adherent on the other side.

And of course, given the statistically significant imbalance of liberals vs conservatives in the academic social sciences, it's not really a surprise that you don't often see research pointing in that direction. ;-)

December 9, 2017 | Unregistered CommenterNiV

I might add that I'm not a proponent for reasons very similar to those expressed by Andy above, and IMO, Andy's counterarguments are important caveats to using the inclusion of experimental data to referee the fight.

December 9, 2017 | Unregistered CommenterJoshua

Dan,

"Thus, if we take Jost’s compilation of studies featuring CRT at face value, his finding that liberals score higher on it is a reason to infer that liberals are more vulnerable, not less, to politically motivated reasoning than are conservatives."

If MS2R and PRMP are both symmetric, then isn't the implication that libs and cons have the same S2 abilities population-wide? In other words, if there is some S2-ability asymmetry, why isn't it making the side with greater S2 abilities have more MS2R in your results, as you imply it should in the above quote?

December 9, 2017 | Unregistered CommenterJonathan

Woops - sorry - that was kind of garbled. What I meant is if overall PMR is symmetric (independent of cognitive abilities/dispositions), and MS2R is also symmetric (the tendency to have more PMR when higher in some cognitive abilities/dispositions), then that seems to imply that somehow the S2 component is a wash, doesn't it? So, either S2 abilities/dispositions that promote higher PMR in MS2R are also symmetric, or there is something else required to make up the difference.

December 9, 2017 | Unregistered CommenterJonathan

@Jonathan--

NOt sure I'm 100% understanding your query. But I'll answer what the question I think you are asking, and then you can tell me what I'm missing.

Yes, if one looks at PMR and MS2R for population as a whole, the symmetric but opposing force of them on right & left balances out, & it looks like the relationship between, say, CRT & perceptions of climate change etc. is zero. But such a model would be hiding that there is in fact a very large effect -- just opposing & symmetric conditional on cultural outlooks (or whatever measure of group identity one is using in the model).

When I said, "... take at face value ...," I meant take JJ's purported measurement of asymmetry in biased reasoning at face value. As I said in the post, the symmetry of PMR and MS2R is a reason to doubt the validity of the self-report measures of cognitie style.

December 9, 2017 | Registered CommenterDan Kahan

Dan,

Just checking: So, you are saying that the relevant components of S2 to MS2R are symmetric, even though CRT is not?

So, one wonders then why a motivated reasoner with high S2 abilities just decides not to use their high CRT to their motivated advantage, or what external effects keep this usage balanced...

Or, are you arguing that CRT must really be symmetric? In other words, symmetric turtles all the way down?

December 9, 2017 | Unregistered CommenterJonathan

@Jonathan-- I think ideological differences in CRT scores are too minuscule (r to affect things as a practical matter.

December 9, 2017 | Registered CommenterDan Kahan

Dan -

FWIW -

Indeed, it’s clear that the main criterion Jost used to flag these particular experimental studies was that they reached a result congenial to his hypothesis. We can tell that he resorted to cherry-picking of this sort because he didn’t cite a single one of the myriad experimental studies that suggested that liberals are as prone to ideologically motivated reason as conservatives.

I am always disappointed when you engage in that sort of (IMO) hyperbole. The basic argument of your critique of Jost is (IMO) interesting and important - but your editorializing about intent (I suppose you might claim plausible deniability) seems (IMO) to extend well beyond the scientific basis of that critique. I am left to wonder why you include, what I consider to be extraneous, and IMO basically juvenile shade-throwing. Just seems completely unnecessary. You can easily point out the relative weakness of Jost's survey without resorting to an accusation of cherry-picking (which generally implies intent).

For me, it reduces the power of your work - as it suggests that you're willing to draw conclusions beyond the supporting data. And further, - given what you write about polarizing messaging increasing polarization - it just seems rather incongruent.

Of course, I'm only one reader our of at least 14 billion, so my opinion is even less than a drop in a bucket.

December 9, 2017 | Unregistered CommenterJoshua

I meant to add that I suppose it's possible that you have other information, beyond what you wrote in the post, that provides a more solid basis for the "shade" as it were - but if so, then writing this post without including that information doesn't really provide a satisfactory explanation for including the hyperbole (IMO).

December 9, 2017 | Unregistered CommenterJoshua

Dan,

If not CRT, what about other non-minuscule asymmetric and apparently S2-correlated properties, such as college education level? Are you expecting these to not show any MS2R correlation? Or, are you expecting instead that the underlying attribute is symmetric, even though the measurement (in this case college education level) is not? And, further, that the property actually measured (the education that college provides) doesn't enhance S2 skills sufficiently to impact MS2R on its own?

What I'm getting at here (if it isn't obvious) is that if PMR and MS2R are both symmetric, then either any S2 properties that could enhance MS2R must also be symmetric, or their asymmetries are all too small even when aggregated together to make a significant difference, or they magically balance (some on right, others on left), or something even weirder is happening...

December 9, 2017 | Unregistered CommenterJonathan

@Joshua-- I took cherry-picking out (or crossed it out). I think Jost picked the experiments to illustrate that his findings are consistent w/ experimental studies--that's not something that betrays bad faith or intentions. But it is necessarily one-sided

December 9, 2017 | Registered CommenterDan Kahan

@Jonathan--

I think I see the point & more or less agree with it. The MS2R studies -- even experimental ones -- report findings that suggest that at particular level of CRT (or science comprehension or numeracy or education) , magnification of biased information processing is symmetric. But if the predictors occur at different frequencies in diverse populations, then there will in the real world be a greater observable quantity of politically motivated reasoning (likewise if issues that excite PMR are asymmetrical in their power to excite a perception of identity threat). But at least for now, I don 't think there is evidence that predictors differ in practically meaningful sense.

December 9, 2017 | Registered CommenterDan Kahan

"But at least for now, I don 't think there is evidence that predictors differ in practically meaningful sense."

OK - I'll let it rest at that. The burden is on the symmetry theorizers (such as yourself) to demonstrate that anytime evidence shows that the predictors differ in practically meaningful ways, they end up doing so in a way that keeps the simultaneous symmetry of PMR and MS2R. Also part of the burden, explain why do they always end up doing so. Personally, I think that's a pretty big burden.

December 9, 2017 | Unregistered CommenterJonathan

Dan -

Fair enough.

December 9, 2017 | Unregistered CommenterJoshua

@NiV
>There are lots of possibilities.... Reasoning characteristics can correlate with political orientation... (/ education / upbringing / etc.)

Which reasoning characteristics? I think this just begs the question. As a first order effect, how we reason is rooted in biology. While one may or may not see some second or third order effects, culture governs *what* we think rather than how; its in-group / out-group criteria are arbitrary and emergent. The competition of emotive narratives and resulting social enforcement of the winners for particular cultures, are common mechanisms across the board; leading to an expectation that biases and consequent motivated reasoning would likewise be the same for any culture.

>Openness to new experiences/knowledge versus respect for tradition. Open people are inclined to like things just because they're new and unfamiliar.

What is new and unfamiliar, and indeed traditional, is relative to cultural norms. As far as I recall Dan's work shows that openness does not vary to a degree that practically matters across separate cultural groups. And this is also an expectation because all cultures beyond the newly minted stage have their orthodox and enforcer branches, their mainstream center, their heretic fringes, their conflicts between the new front end evolutionary edge and their core protected narrative dna, etc. They are dynamic and diverse and evolving entities that each encompass, and indeed need, many shades of contribution. They are not static fixed entities that have only one sort of people within, and indeed cultures may exchange individuals, whose reasoning characteristics travel with them, even though they will then apply them to different goals.

> 'Hierarchical' respect for authority versus 'individualist' liking for the maverick outsider. Some people automatically trust big institutions, impressive credentials, age and experience. They assume the standard way of doing it is most likely correct. Others believe in intellectual revolutions, new discoveries overturning old dogmas, and continual change, and tend to automatically distrust Establishments and authority figures.

It must be borne in mind that the core narratives that are typically perceived as the main characteristics of any culture, are all false. They are just a flag to rally around and the means to create an enforceable consensus, which in turn means a sizeable social block acting in concert. For example the CAGW flag, i.e. the dominant narrative of a *certainty* of imminent (decades) catastrophe. These flags are nodded to, but all those sorts of people you list exist in all cultures, and there is miniscule statistical leaning towards any particular characteristics which the flag advertises or implies. This is true of the Con flag, the Lib flag, religious flags, the CAGW flag, all of them. This is why, as Dan notes very frequently here, when culture gets entangled in an issue in a big way, it is not the 'what' that mostly counts (therefore including unbiased reasoning regarding the 'what'), but the 'who'. Not to mention that authority is also relative to culture. And while newer cultures can themselves overturn older ones, causing more statistical differences at the boundary times, individuals of all sorts can be pretty quick to shift interests when a tipping point is reached.

>'Empathic' reasoning from consequences versus 'cold' logic. Some people will shy away from a line of reasoning if it leads to unpleasant or immoral consequences for people they sympathise with.

Absolutely, but the determination of 'immoral' springs from culture, and 'unpleasant' may be likewise, unless it is something much more basic going on with the people sympathized with. This will happen to all culturally immersed folks no matter what the culture is, and yet if the cultural determination does not indicate an actual or likely immoral or unpleasant outcome, they will be much happier with logic. This is not different per culture.

>System 1 and System 2 processing. Some people always go with their initial 'gut' impression, others are more inclined to to process questions analytically and check the results.

Also the same regarding all cultures. The 'gut' thinking will work via the invocation of instilled cultural norms, but the *way* it works is not different per culture.

>Analytic versus synthetic reasoning / Simple single-step versus complex multi-step causality / Concrete versus abstract reasoning

What evidence do you have that specific cultures would have a significantly different weighting of such people, to a first or even second order degree? The way in which cultures emerge and are maintained, suggests that all such people would be found in all mainstream cultures (of course when you get down to some extreme fringe pseudo religion with 30 people in it or whatever, statistical distribution fails, anything can happen with such low numbers).

>Trusting friends versus trusting outside experts. Some people base their trust on familiarity and long experience with people they know well, and distrust strangers. Others vice versa.

Trust does have a weighting towards culture, very much so if the topic invoking the necessity of trust involves the main domain that the culture occupies. However, this weighting has no dependency on what the culture actually is, only that there is adherence to one. I never heard of people who trust strangers more than friends, unless maybe they have no real friends so it's not truly valid. But if this is common why would there be more of them who are cons rather than libs, or Catholics rather than Protestants, or whatever other cultures whichever way around; they would very likely trust an in-group stranger more than an out-group one, but that again is not a different expectation per culture.

"There are some policy positions that are factually and morally *obvious*, to the point where it's inconceivable that any rational person could disagree with them. But those cultural outsiders over there don't believe in them. Therefore those people cannot be rational, and it's just a matter now of figuring out what specifically is wrong with their reasoning ability."

Indeed. People often can't see past the norms of their culture. But for sure there are widespread examples of this for any strong culture; if you think this PMR stance is actually significantly worse in some cultures than others (and have reasons to think so), what are these, and as I asked above, what could possibly be an underlying (evolutionary) causation for such?

December 9, 2017 | Unregistered CommenterAndy West

I will offer the following ramble, Andy, So I don't have to put IMO in every sentence, I'll put out one big IMO right here at the beginning.

If you happened to watch the video that Jonathan linked downstairs, of a discussion between Haidt and Pederson where they discuss patterns of association between personality styles and ideological orientation, they spoke of a kind of Yin and Yang pattern where people play oppositional behavioral roles that form a balance that results in a kind of societal stability.

I'm not sure that sort of construct works, exactly, across different cultures or across a time span that extends deeply prior to our own - but since I've watched that video I've been thinking a lot about something that I've been ruminating about for a while now; basically that there is a Yin Yang (for a lack of a better expression) that works to describe a general pattern in my society during my lifetime. And while I don't think that it really correlates with different forms of cognition or even values, I do think that it associates, at least to a meaningful extent, with broad-scale ideological divisions.

Roughly speaking, I think of it as a balance between the "disciplinarian father" and the "nurturing mother." Within a family unit, the mother and father often assume those roles to provide a broad and healthy environment for the child. (Most often they might each play a role that is more or less mutually exclusive with the other parent playing the opposite role, or they may, less commonly, each establish a balance between those roles within themselves internally).

It certainly doesn't have to be aligned by gender, and indeed in many families the father can be the nurturer and the mother the disciplinarian. I think that the roles are more along the lines of a Jungian archetype than anything that has some causal connection to any particular characteristics such as mother/father or liberal/conservative. And indeed, at times parents can switch in those roles just as, indeed so can liberals and conservatives depending on the particular issue at hand - e.g., right now liberals are trying to play the role of tough and strict authoritarian w/r/t sexual harassment whereas conservatives are trying to play the role of understanding and flexible judge (It's ok for a 30 year old Roy Moore to date teenagers because it's common in the South and wasn't considered so unacceptable in previous decades). Consider how libz and conz might switch in those roles if they are considering the norms of how girls are treated in Muslim countries (libz= we can't judge their culture and conz= Muslims are undisciplined barbarians that need to be dealt with authoritatively).

The "mother/father" frame is only one of many of such "archetypal tensions" that form a balance in society as a whole, and while it's association with ideological identities can shape-shift in various contexts, IMO, that particular archetypal tension does seem to play out in general as fairly foundational and central tension that plays out as a libz=nurturer and conz=disciplinarian framework.

I want to be clear that I don't think that "evolution" "explains" that lib/con pattern of association (not the least because I almost always think that reverse engineering the mechanics of evolution is almost always just so storyfying). Nor do I think that a difference in "values" explains that pattern (I see values as crossing those boundaries, i.e,, a disciplinarian's goal is nurturing and a nurturer's goal is to teach discipline). And I certainly don't think that the pattern is particularly associated with an asymmetry in the mechanics of reasoning so much as with a outcome of a shared mechanical process where there is no real asymmetry (although there is no shortage of people who seek to reverse engineer from that asymmetry in outcome to impute a mechanical difference in how people reason).

So it can look very much like an asymmetry in reasoning, or as per this post an asymmetry in degrees of "motivated reasoning" (e.g., one frame might be that disciplinarians are more deliberate and analytical in considering the effect of actions and nurturers are more instinctive and intuitive in their expectations of outcomes from actions). But in reality, the asymmetry is only like a very variable movie screen onto which people can project their just so visions.

And I would guess that by definition, there should be change over time or place in which "archetypal tensions" might rise in the immediacy of importance (i.e., in some cultures, there might be reasons why a tension between "loyalty and independence archetypes" would rise in prominence - I think of a shift I felt in that regard in my lifetime when I lived in Asia for 1.5 years).

Like I said, not sure if ANY of that makes any sense, but I wanted to get it down just so I can try to crystallize some of my thoughts. Hopefully I won't be too embarrassed when I read through it again tomorrow.

December 9, 2017 | Unregistered CommenterJoshua

@Jonathan--

<<The burden is on the symmetry theorizers (such as yourself) to demonstrate that anytime evidence shows that the predictors differ in practically meaningful ways, they end up doing so in a way that keeps the simultaneous symmetry of PMR and MS2>>

Burdens of proof make sense in legal proceedings (conditional on legal proceedings making sense; I'm not sure they do), but not really for empirical scholars. Legal proceedings come to closure on facts. But empirical scholarship is aimed at generating material that has some evidentiary weight in permanent process of adjudicating competing hypotheses. Accordingly, anyone who contributes anything to empirical assessment bears the burden to establish that what he or she is adding to scholarly discussion has a likelihood ratio ≠ to 1.

December 10, 2017 | Registered CommenterDan Kahan

"As a first order effect, how we reason is rooted in biology. While one may or may not see some second or third order effects, culture governs *what* we think rather than how;"

I think a lot of the how is also rooted in culture - the scientific method, for example, took centuries to develop and has to be learned.


"What is new and unfamiliar, and indeed traditional, is relative to cultural norms."

True. But whether you depend on tradition or your own resources might not be.

"As far as I recall Dan's work shows that openness does not vary to a degree that practically matters across separate cultural groups."

Yes. And he did the test precisely because it had been proposed as a mechanism for the differences.

"For example the CAGW flag, i.e. the dominant narrative of a *certainty* of imminent (decades) catastrophe"

Actually, the narrative I'm thinking of here is the "Peer reviewed science, 97% of scientists, expert climate scientists" narrative, versus the 'Galileo effect' on the sceptic side. Believers (so the sceptic narrative goes) put too much blind trust in authority.

"These flags are nodded to, but all those sorts of people you list exist in all cultures, and there is miniscule statistical leaning towards any particular characteristics which the flag advertises or implies."

I'm not proposing that these are the actual explanation. You asked what possible mechanisms AT adherents could propose, and I've listed some that they have or could. You have to do the statistics to eliminate them. For some of them, they've already been proposed, tested, and rejected. For others, I'm not so sure. But my main point is that there is an endless list of hypotheses that can be constructed - we don't know enough about the mechanisms yet to be able to eliminate them all.

"Absolutely, but the determination of 'immoral' springs from culture, and 'unpleasant' may be likewise, unless it is something much more basic going on with the people sympathized with."

I didn't say otherwise. The hypothesis is that some people are more/less influenced by it, not that it's not based on culture.

And for that matter, given that I said above that it may be culture causing the reasoning characteristics, as well as reasoning characteristics causing culture, it wouldn't make any difference to the AT if it was.

"The 'gut' thinking will work via the invocation of instilled cultural norms, but the *way* it works is not different per culture."

How do you know?

"What evidence do you have that specific cultures would have a significantly different weighting of such people, to a first or even second order degree?"

What evidence do you have that they don't?

I'm generating hypotheses, not reporting results. (More specifically, I'm demonstrating that generating hypotheses is easy. No more.)

December 10, 2017 | Unregistered CommenterNiV

@NiV

I agree that if you take 2 cultures very far apart, e.g. an ancient hunter gatherer society and a modern US culture, say, you would hope, even after adjusting the tests to be appropriate to each culture, that some advance in restraint of PMR might be observable. If so there is a plausible cultural evolutionary reason for this, i.e. the slow maturation of more reasoned algorithms like the scientific method, as you note. But as we know science is still very fragile to cultural bias, and nor do many people in society take a very active part in these algorithms, so maybe the advance isn't as much as we'd like to think. (And likely an individual baby robbed from the older culture and brought up in the modern one, would behave just like a modern person). However for any mainstream cultures in the modern US, such as those Dan typically measures and indeed as I exampled by name, no such cultural evolutionary gulf exists anyhow.

>How do you know? (on gut thinking)

Because cultural bias / polarization has been shown, e.g. on this very blog, that does not look different in nature per culture, only different in *what* they are biased about.

>What evidence do you have that they don't? (nuanced reasoning modes)

I do not, and maybe no-one has measured down to this low-level resolution. But anecdotally of all the diverse people you have experienced who are Libs or Cons or Catholics or whomever from other cultures you have met, are you really going to expect a primary weighting of such narrow modes to particular cultures, considering also that more generic modes, per above, have been measured and found (assuming the cage match result here) to be symmetrical, plus...

>I'm generating hypotheses...

Generating hypotheses is fine and dandy, but the deeper point here (I should have expressed this directly!) is that none of these possibilities you raise, address my original question, i.e. even as hypotheses for which some at least are unmeasured, so unknown. Because they just defer the issue by one stage. So, what possible causal evolutionary reason (as in a plausible proposal is fine, not a proof) could there be for these various particular modes to have become strongly culturally dependent.

December 10, 2017 | Unregistered CommenterAndy West

"But anecdotally of all the diverse people you have experienced who are Libs or Cons or Catholics or..."

This is ambiguous. I meant diverse *within* each culture.

December 10, 2017 | Unregistered CommenterAndy West

@Joshua

Assuming I understood this right, I think you have a valid perception here that merits thought and explanation.

I think this perception is readily glimpsed in terms like 'Fatherland' and 'Motherland', and various other cultural icons such as the American Eagle or images of the Madonna or whatever, which may at various eras be either weak / general or strong / specific identifiers for associated cultures (Fatherland has obviously been a very strong identifier in a recent era in Europe), and indeed including national and group popularly distinguished 'personas' such as disciplinarian Cons and nurturing Libs, as you note for these modern cultures within the US.

And along the way of adding my twopence worth, I agree that typically, evolutionary principles cannot tell us anything about particular cultural characteristics, e.g. for Libs or Cons. Notwithstanding the caveat that if 2 cultures are very far apart in cultural evoluionary terms, per the answer to NiV above, so one can start to make some generalizations based on differential development across such a gulf, cultures are emergent, and their particular identity characteristics come from the emergence not from the rule system within which the emergence occurred. The advantage of referring back to this rule system, is that because it developed over many millions of years (some animals have culture), and in the latter tens of thousands of years at least, across many many thousands of cultures whose diversity dwarfs, for example, all the mainstream modern cultures in the US today, it sets a strong expectation of *constant* rules. I.e. this system, developed over such a long period of gene-culture co-evolution, is highly unlikely to be different (in a primary way) for any cultures that are closish on the overall topography, such as US Libs and Cons, or indeed Catholics, despite this culture spans millennia (all cultures adapt to keep up, or alternatively they wouldn't be here anymore!) What is different in each case is the emergence , and hence the dominant narratives and characteristics for each culture, which also influence each other (e.g. via alliance, or backlash, heretical split, and as you note, oppositional balance). And the rules also tell us therefore, not what will be different or specific about Libs, Cons, Catholics, etc, but what will be the same, e.g. they will all be biased, because bias is at the heart of what a culture is.

The diverse example scenarios you give (Muslims, teenagers, sexual harassment etc), plus that in practice a group sometimes seems to be effectively working in reverse of its main advertised image regarding a particular issue (e.g. Dems being authoritarian on harassment), and especially regarding your note about the 'shifting over time' of attitudes and tensions associated with these examples, emphasizes the practical reality that these advertised cultural personas are largely illusory. They represent a flag to rally around much more than they represent a practical way of living, the latter of which needs huge diversity and compromise and adaptation anyhow. While the flag has to be nodded to by all immersed adherents, meaning one should pick up some minor statistical differences in behavior, as Dan does, the huge prominence of those flags in society (especially in a more tribal phase, like now in the US) disguises the fact that each culture is fully staffed by the full panoply of human behaviors (e.g. some you mentioned like nurturing, disciplinarion, and many more) and as a first order effect across each of their whole populations, they all think in the same (diverse) ways. As Dan's work shows, even en-masse with quite simplistic tests, what they really think when one tucks the flag away, diminishing identity challenge, is quite different to the advertised narrative.

Incidentally, I think this all makes it far more complex to sanction / change societies (typically more entrenched in the past) that have practices we may consider distasteful or abhorrent. A society practicing mutilation for instance, will have no less caring mothers than ours, but what it means to be a caring mother in their emergent narratives, is different to what that means for our emergent narratives.

At any rate, while the rise of a conservative society may beg a liberal one in opposition, or the absolute dominance of a major religion may invite a heretical split to restore some balance at the societal level, I don't think I would view disciplinarian / nurturing or similar specific characteristics as 'foundational'. Though for sure they are important touted flag elements, part of the emergent narratives of particular groups, they are not part of the rules. And so like all flag elements they don't reflect the true diversity beneath. Such narratives and associated cultures have been astonishingly different in the past, and among those that have particularly stuck me are the (typically deity based) duality personas - nurturing and terrible both at the same time - which have faded in modern times, and complex yet very different cultures like that of the Spartans. In particular conflict topics where cultures face off and both wave their flags furiously, behind which facts fade into a lost horizon, there can indeed *appear* to be differences in PMO one way or another. But the advantage of Dan's great investigation is that it can look beneath the flags (largely - tests that completely eliminate all bias effects in the data gathering are hard, nor do I think Dan is distanced from all priors for the analysis stage, but then who is), and to date at least the evidence weighs very strongly in favor of PMO not being dependent on culture.

December 10, 2017 | Unregistered CommenterAndy West

"But as we know science is still very fragile to cultural bias, and nor do many people in society take a very active part in these algorithms, so maybe the advance isn't as much as we'd like to think."

Dunno. They are on a very short list of shibboleth issues, like global warming, gun control, or evolution. But how much of science is that?

"Because cultural bias / polarization has been shown, e.g. on this very blog, that does not look different in nature per culture, only different in *what* they are biased about."

So do you think an AT adherent would propose that as a mechanism?

"But anecdotally of all the diverse people you have experienced who are Libs or Cons or Catholics or whomever from other cultures you have met, are you really going to expect a primary weighting of such narrow modes to particular cultures, considering also that more generic modes, per above, have been measured and found (assuming the cage match result here) to be symmetrical"

Anecdotally, I think the line of reasoning I described above is very natural and very common.

But I think most people who hold the belief don't reason that far. The basic argument goes: "There are some policy positions that are factually and morally *obvious*, to the point where it's inconceivable that any rational person could disagree with them. But those cultural outsiders over there don't believe in them. Therefore those people cannot be rational, and it's just a matter now of figuring out what specifically is wrong with their reasoning ability." The fact they haven't done so yet doesn't invalidate their basic argument.

"Generating hypotheses is fine and dandy, but the deeper point here (I should have expressed this directly!) is that none of these possibilities you raise, address my original question"

That question being: "I've not delved into this topic, but I've wondered on occasion what on Earth causal reason AT supporters might propose that could support their position"?

"Because they just defer the issue by one stage. So, what possible causal evolutionary reason (as in a plausible proposal is fine, not a proof) could there be for these various particular modes to have become strongly culturally dependent."

Who said there was an evolutionary reason?

Or local community centre has two clubs: the chess club, and the 'drunken whack-a-mole' club. The two groups have very different cultures. One values intellectualism, logic, precision, and thinks chess is the best game ever. The other culture values partying, getting drunk, and smashing things with hammers. Each thinks the other's opinions on such questions as "What's the best game?" and "Does adding up the scores correctly really matter?" are correlated with their personality and mental style. Either intellectuals are attracted to chess, or liking chess makes them intellectual, or they have a common cause.

What evolutionary reason could there possibly be for chess club membership to be correlated with intellectualism? For most of human evolutionary history, chess didn't even exist! I don't think it's hard to figure out why there might be a correlation between group membership and personality, in the case of a chess club. I don't think it's blindingly obvious that we should expect there not to be one with respect to politics.

December 10, 2017 | Unregistered CommenterNiV

Andy -

Thanks. I'll spend some time with your response later. Just wanted to say now that on reflection, I do think that I'm trying to have my cake and eat it too; one the one hand I want to say that the that father/mother archetypal balance is foundational and universal, and then on the other hand I want to say that it is a function of time period and place. Hmmm.

December 10, 2017 | Unregistered CommenterJoshua

Dan -

Again, FWIW,

I'm glad to see that you went further and struck out the rest of that paragraph - based on John's reaction. I think that the post (and your argument) is improved by having done so.

But I did want to say, again, it seems to me that the value of your analysis lies in pointing out that you find Ditto's survey to be more informative by virtue of a less circumscribed evidence dataset, and Jost's problematic methodology (because of a sub-optimal explanation of the rational for the inclusion/exclusion criteria used).

Going beyond a more tightly controlled critique, IMO, seems gratuitous. But in the end, interestingly, I think that in your putting that material in and then removing it, I wind up better informed than if it had never been there to begin with. Perhaps, even, the strength of the dialog between scientists is strengthened by the openness to and accountability on the part of people who have disagreement. And I think that being open to and reacting to John's feedback is a reflection of your focus on evidence-based analysis.

December 10, 2017 | Unregistered CommenterJoshua

@NiV

>Dunno. They are on a very short list of shibboleth issues, like global warming, gun control, or evolution. But how much of science is that?

I think, or at least hope, that a very large amount slips under the radar. But considering it is precisely the issues people purport to care about most that rise above the cultural radar, I'd for sure regard this as still a major issue in modern times. This has also happened throughout history, so a large array of science issues now accepted were only after years or decades or even centuries of cultural resistance, which in some cases (especially medical issues) were highly damaging in the interim. Still the case very recently with the saturated fats consensus, now collapsed; likely the health of hundreds of millions of people adversely affected, some seriously. As the profile of various science issues has been raised in recent times due to more money and kudos associated with this endeavor, this ironically brings yet more issues into the cultural radar, and so far they are generally proving fragile to its attention.

>So do you think an AT adherent would propose that as a mechanism?

I don't understand what you mean here. How can something whose nature is the same for every culture, be used as potential support for AT?

>Who said there was an evolutionary reason?

As noted above, the system of culture is derived from very long evolutionary development. It didn't spring up with modern cultures such as the libs or dems, or even millennia old cultures like the Catholics. A vast array of differing cultures have been supported on mechanisms that thus far, biological and cultural evolutionary understanding has not required to be culture specific for its explanations. So if there isn't an evolutionary underpinning for why different cultures might systemically support significantly different modes of thinking, one has to seriously question whether any such dependency, as a main effect at least, could exist. And conversely, if one is seriously proposing that such an asymmetry does exist, one ought to be marshaling candidates for such an underpinning pretty quickly, because this will certainly be needed for survival of the hypothesis.

>What evolutionary reason could there possibly be for chess club membership to be correlated with intellectualism?

As noted to Joshua above, evolution tells us nothing about the specific characteristics of what one culture values over another, e.g. their main narratives and touted reasons for existence. These are properties of emergence. Evolution determines the rule system in which the emergence occurs. But in any case, neither your chess club or your whack-a-mole club are cultures. They are clubs formed for minor diversionary purposes possessing no significant responsibility for the main cultural narratives in the society in which they are embedded, no existential clauses, no significant leverage on social infra-structure, likely v small memberships (per above, anything can happen with very small numbers) etc etc etc. This doesn't mean that clubs will necessarily feature zero cultural effects, albeit selection for them will be very weak. For instance they may develop some group think on the leadership committee, which if it occurs in both your clubs, will occur via the same mechanisms, even though what beliefs the group think promotes will be different in each case.

December 10, 2017 | Unregistered CommenterAndy West

Dan,

"Burdens of proof make sense in legal proceedings (conditional on legal proceedings making sense; I'm not sure they do), but not really for empirical scholars."

OK - I'll try a different explanation. Your data suggests that there are now four different scales that all have the same center point: the political left-right scale, PMR, MS2R, and SC. Note as well that the political left-right scale is different in other countries (such as Western Europe).

Doesn't it seem that having 4 such psychological variables that are all uncorrelated with respect to each other, even though three are related to a particular type of cognition and the fourth is often the object of that cognition, is a bit magical? What could be the explanation?

December 10, 2017 | Unregistered CommenterJonathan

"I think, or at least hope, that a very large amount slips under the radar. But considering it is precisely the issues people purport to care about most that rise above the cultural radar, I'd for sure regard this as still a major issue in modern times."

Again, I don't know about that. I think a lot of stuff, like technology, medicine, engineering and so on, that I think people agree on, and care about. The circumstances that entangle scientific issues in politics are complex, but quite unusual.

"I don't understand what you mean here. How can something whose nature is the same for every culture, be used as potential support for AT?"

What I mean is, you asked for what evidence an AT adherent would propose in support of their hypothesis. Why would an AT adherent cite evidence that contradicts their hypothesis?

"As noted above, the system of culture is derived from very long evolutionary development."

Yes, but why should the reason for "these various particular modes to have become strongly culturally dependent" be evolutionary? Why not historical accident, or the differential attractiveness of ideas, or differential comprehension?

"But in any case, neither your chess club or your whack-a-mole club are cultures. They are clubs formed for minor diversionary purposes possessing no significant responsibility for the main cultural narratives in the society in which they are embedded, no existential clauses, no significant leverage on social infra-structure, likely v small memberships (per above, anything can happen with very small numbers) etc etc etc."

Hmm. You're evidently using a different definition of "culture" to me. We may be talking at cross purposes.

December 10, 2017 | Unregistered CommenterNiV

@NiV

>Why would an AT adherent cite evidence that contradicts their hypothesis?

I presume they wouldn't. But I never implied otherwise. In truth I've completely lost the plot as to what you mean here. I looked back up-thread to trace where one or both of us must have misunderstood intent, but it's still no clearer to me.

>Yes, but why should the reason for "these various particular modes to have become strongly culturally dependent" be evolutionary? Why not historical accident, or the differential attractiveness of ideas, or differential comprehension

Narrative information, which includes ideas, is indeed differentially attractive, and is selected on this basis whether arising from accident or design or a mixture of both. Higher selection value in a cultural context is granted not by veracity but by emotive impact, which maximizes the idea's penetration and re-transmission to other individuals. While there are other characteristics that help with selection value, none of these imply that the recipients (i.e. people) are different in any fundamental way. The mechanics work the same no matter what culture may emerge at any time and place from an iterative co-evolutionary selection of narratives, and so which cultures the people happen to belong to or not. Within each particular culture is a very diverse set of people, but their span of reasoning abilities and styles (to a 1st order at least), will look the same (if the cultures are all near enough on the overall landscape, e.g. all the modern mainstream cultures in the US). What defines them is the set of narratives they each believe in.

Yet if one proposes that emergent cultures are not just the result of iteratively selected narratives, but that they are composed (as a first order or at least major characteristic) of people who actually reason in different ways, this implies a much more fundamental difference between cultures; potentially more of a hard-wired thing, so to speak. Given the vast range of cultures both current and historical, their rises and falls (sometimes swift, sometimes glacial), their frequent exchanges of individuals, their fluid and constant morphing, then proposing how asymmetrical modes of thought (e.g. per the head post, some cultures supporting significantly less PMR than others), can align to them, i.e. invoking this much more fundamental characteristic, requires causal justification. Not least also because it flies in the face of pretty much all cultural evolutionary thought as far as I know. A culture practically is an enforced consensus which is necessarily untrue. Hence its job, so to speak, is to create bias (for the culture), so one that creates less bias for the same degree of cultural immersion, would not last long in competition. At any rate, an explanation must be rooted in evolutionary theory, because that's where the theory of culture is itself rooted. The current position is that 2 people from different cultures are both similar human beings who believe in different narratives. The asymmetrical proposition is instead, that 2 people from different cultures are actually different human beings. Both these ignore all sorts of caveats, but that's the kind of distilled clinical essence.

Culture has many definitions in our language unfortunately. Yet for sure I'm not talking about a night at the opera or the chess club, but culture as comprehended by the discipline of cultural evolution, i.e. the mainline cultures that drive society: strong political movements, religions, etc. which have well-mapped characteristics.

December 10, 2017 | Unregistered CommenterAndy West

@Jonathan

I must have been missing a lot in reading previous comments-- I had no clue you were going to where the most recent ends up.

1. The cultuaral scales are uncorrelated by design.

2. It's certainly *not* the case that they are uncorrelated w/ left-right ideology.

3. They are centered at their *own* midpoings b/c they are normalized.

4. uncorrelated variables can still have different correlations w/ a 3d variable.

5. Now what trap have you set for me?

December 11, 2017 | Registered CommenterDan Kahan

Dan,

Sorry - after sleeping on it, I realized I was mistakenly thinking of "uncorrelated" as if it were a transitive property. That was greatly magnifying my surprise.

The only surprise I have left is that there are S2 measures that are asymmetric (like analytic reasoning, CRT and education level), yet MS2R and SC are by your measures symmetric.

I guess my question comes down to:
If some self-sorting process is going on in western societies that either pushes high S2'ers somewhat left or enhances S2 somewhat when on the left, then why doesn't this happen to MS2R or SC?

December 11, 2017 | Unregistered CommenterJonathan

Link drop - van der Linden, Leiserowitz and Maibach have a new study that says preaching the consensus works, as long as you hide the message like spinach in a brownie:
https://phys.org/news/2017-12-facts-consensus-bridges-conservative-liberal-climate.html

However, once the treatment group were exposed to the 'social fact' of overwhelming scientific agreement, higher-educated conservatives shifted their perception of the scientific norm by 20 percentage points to 83% - almost in line with post-treatment liberals.

I can't find a non-paywall version.

December 11, 2017 | Unregistered CommenterJonathan

W/r/t that link from Jonathan -

Moreover, the latest research confirms the prior finding that climate change scepticism is indeed more deeply rooted among highly educated conservatives. Yet exposure to the simple fact of a scientific consensus neutralises the "negative interaction" between higher education and conservatism that strongly embeds these beliefs.

Seems to me like a really odd paragraph. I am curious (would one say "scientifically curious?") which highly educated conservatives haven't already heard the line that "97% of climate scientists have concluded..." Just who are those highly educated "conservatives" that have somehow not heard that message before?

As much as I have seen this "gateway effect" argued, and the accompanying statistics used to argue about the "consensus gap," I have yet to see any quantification of the number of people who haven't heard the "consensus"message. I would think that at this point that number must be low, and that closing the "consensus gap" isn't so much a matter of delivering that message to people who haven't heard it before.

And it does always bug me when they conflate "global warming is happening" with "global warming is happening and it is largely driven by ACO2 emissions." It isn't that I doubt that many "conservatives" believe that climate change isn't happening; despite activist "skeptics" trying to (conveniently, and employing a double standard with respect to the "appeal to authority fallacy) assert something on the order of "Hardly anyone doubts that the climate is changing or that there is a GHE effect, we only question the magnitude of that effect." evidence such as that presented by Dan shows that simply isn't the case.

And once again, the problem with studies such as that one is that it doesn't really account for real world messaging, where people hear the messaging (and counter-messaging) from ideologically-aligned sources. IMO, the chances of "external validity" is prolly pretty low.

December 11, 2017 | Unregistered CommenterJoshua

Andy -

I'm not sure I get the link to "Fatherland" and "Motherland," but your comment at least suggests that you got what I was going for well-enough to make me think it wasn't complete nonsense.

Notwithstanding the caveat that if 2 cultures are very far apart in cultural evoluionary terms, per the answer to NiV above, so one can start to make some generalizations based on differential development across such a gulf, cultures are emergent, and their particular identity characteristics come from the emergence not from the rule system within which the emergence occurred.

Yes. But I happen to think that it is very difficult, even, to find some objective basis on which to determine the relative importance of differences. Just as environmental influence can make it ambiguous as to whether differences are rooted in some fundamental rule of evolution as opposed to an emergent phenomenon, so can environment influence what line of criteria one uses to measure difference (or to rank "differences"), and in a circular way, influence what criteria one uses to measure those differences. For example, while many Asians tend to view Americans as rude (generally speaking), when I was living in Korea I was frequently surprised by the indifference shown to the norms of politeness I was accustomed to (for example, whether people exchange pleasantries with a shop clerk, or whether one asks a stranger whether they're married when first meeting them, I could provide a very long list of such).

The advantage of referring back to this rule system, is that because it developed over many millions of years (some animals have culture), and in the latter tens of thousands of years at least, across many many thousands of cultures whose diversity dwarfs, for example, all the mainstream modern cultures in the US today, it sets a strong expectation of *constant* rules. I.e. this system, developed over such a long period of gene-culture co-evolution, is highly unlikely to be different (in a primary way) for any cultures that are closish on the overall topography, such as US Libs and Cons, or indeed Catholics, despite this culture spans millennia (all cultures adapt to keep up, or alternatively they wouldn't be here anymore!)

I generally agree - perhaps with a caveat as per above. The notion of determining "closeness," seems to me to be to some degree a matter of perspective, largely based on the embedded norms of the observer, and thus vulnerable to the influence of subjectivity.

What is different in each case is the emergence , and hence the dominant narratives and characteristics for each culture, which also influence each other (e.g. via alliance, or backlash, heretical split, and as you note, oppositional balance). And the rules also tell us therefore, not what will be different or specific about Libs, Cons, Catholics, etc, but what will be the same, e.g. they will all be biased, because bias is at the heart of what a culture is.

Again, I agree. I often raise the question, in these discussions in these threads, when differences between groups are being discussed, whether those "differences" are truly meaningful if we consider how the difference between the mean characteristics of one group versus the other compares to the differences within the individual groups.

The diverse example scenarios you give (Muslims, teenagers, sexual harassment etc), plus that in practice a group sometimes seems to be effectively working in reverse of its main advertised image regarding a particular issue (e.g. Dems being authoritarian on harassment), and especially regarding your note about the 'shifting over time' of attitudes and tensions associated with these examples, emphasizes the practical reality that these advertised cultural personas are largely illusory. They represent a flag to rally around much more than they represent a practical way of living, the latter of which needs huge diversity and compromise and adaptation anyhow. While the flag has to be nodded to by all immersed adherents, meaning one should pick up some minor statistical differences in behavior, as Dan does, the huge prominence of those flags in society (especially in a more tribal phase, like now in the US) disguises the fact that each culture is fully staffed by the full panoply of human behaviors (e.g. some you mentioned like nurturing, disciplinarion, and many more) and as a first order effect across each of their whole populations, they all think in the same (diverse) ways. As Dan's work shows, even en-masse with quite simplistic tests, what they really think when one tucks the flag away, diminishing identity challenge, is quite different to the advertised narrative.

Agree without comment.

Incidentally, I think this all makes it far more complex to sanction / change societies (typically more entrenched in the past) that have practices we may consider distasteful or abhorrent. A society practicing mutilation for instance, will have no less caring mothers than ours, but what it means to be a caring mother in their emergent narratives, is different to what that means for our emergent narratives.

Yes, well, that is where the rubber meets the road. Complex indeed.

At any rate, while the rise of a conservative society may beg a liberal one in opposition, or the absolute dominance of a major religion may invite a heretical split to restore some balance at the societal level, I don't think I would view disciplinarian / nurturing or similar specific characteristics as 'foundational'. Though for sure they are important touted flag elements, part of the emergent narratives of particular groups, they are not part of the rules. And so like all flag elements they don't reflect the true diversity beneath. Such narratives and associated cultures have been astonishingly different in the past, and among those that have particularly stuck me are the (typically deity based) duality personas - nurturing and terrible both at the same time - which have faded in modern times, and complex yet very different cultures like that of the Spartans. In particular conflict topics where cultures face off and both wave their flags furiously, behind which facts fade into a lost horizon, there can indeed *appear* to be differences in PMO one way or another. But the advantage of Dan's great investigation is that it can look beneath the flags (largely - tests that completely eliminate all bias effects in the data gathering are hard, nor do I think Dan is distanced from all priors for the analysis stage, but then who is), and to date at least the evidence weighs very strongly in favor of PMO not being dependent on culture.

PMO?

So I'm still struggling with the question of whether the "nurturing/disciplining" tension has some primary place among a lot of such tensions which, I suspect, play out in all cultures (as well as within individuals). One example might be "individuality vs. social consensus" as it plays out in contrasts between "flagged" groups within one society, or when comparing two societies such as seen in the difference between Americans and Asians, generally, w/r/t fulfilling individual needs versus conforming to group norms.

My sense is that these tensions are connected to and embedded in human psychology - for example, as seen in the infant's irreconcilable drives for both independence and connection. And what we see as conflicts in society reflects a kind of desire to seek some closure for the unknowableness of where to find that balance (ambiguity is very hard for people to accept). In couples, that can play out as the father resenting the mother for being too nurturing and the mother resenting the father for being too strict, when what is really playing out is self-doubt for each of the parents as to how to strike the balance and fear that they, themselves are out of balance (in what is always, a somewhat ambiguous situation, that largely depends on a context, a context which is always dynamic and changing).

I see this a lot in Internet flame wars. One example, close to the heart of the climate wars, might be how people polarize around the issue of "appeal to authority" - where in fact, the importance of authority does not conform to a binary conclusion. And so people stake out a polarized claim to one side of the balance, and demonize those on the other side as being out of balance in the other direction, and often, even switch sides of that balance depending on the particular context, which is in the end only a reflection of the fact that the two sides must remain in tension because ultimately, there is no real answer as to "which is better." There are so many which I've seen take the same form (e.g., are oceans becoming more acidic, is Muller a skeptic, do we measure temperatures or estimate temperatures, etc.).

December 11, 2017 | Unregistered CommenterJoshua

Joshua,

" Just who are those highly educated "conservatives" that have somehow not heard that message before?"

I don't see any indication in the phys.org article that says they tested this message on cons who hadn't heard it before. My reading was that this study is showing an almost subliminal tendency to move one's beliefs toward messaging that is presented in somewhat hidden ways. I'm not sure it is a victory for consensus messaging, unless they also show that the effect is robust and not temporary. Also, how much does the content of the message matter? Might a non-consensus message, such as that the moon landing was faked, also show increased belief when presented this way?

December 11, 2017 | Unregistered CommenterJonathan

Jonathan -


I don't see any indication in the phys.org article that says they tested this message on cons who hadn't heard it before.

My point being that they are contending that consensus messaging would make a difference in the real world. And to know whether that is true, then the would need to test for whether the conz have heard it before.

If people have already heard consensus messaging, then even if the effect they find is true and externally valid, what would the impact be in the real world? One argument might be none - because the effect has already played out (which could also mean that the effect has been positive in the real world even if there remains a "consensus gap" - because the real world positive effect was to mitigate counter messaging; w/o the consensus messaging even more of the public would reject that there is risk posed by ACO2 emissions).

I'm not sure it is a victory for consensus messaging,...

Well, yes.

The nature of the study was hidden by claims of testing random media messages, with the climate change perception tests sandwiched between questions on consumer technology and popular culture messaging.

This, it seems to me, is a pretty contrived experimental condition - as people don't typically hear "consensus messaging" in that context, and they are also constantly bombarded with counter messaging. As such, I don't know that the finding is a victory for consensus messaging in the real world.

December 11, 2017 | Unregistered CommenterJoshua

BTW - reading through Joshua and Andy's discussion - has reminded me of a book:
The Honor Code: How Moral Revolutions Happen by Kwame Anthony Appiah. Anyone else familiar with it?

December 11, 2017 | Unregistered CommenterJonathan

Haven't read it, Jonathan. I'll check it out. Thanks.

December 11, 2017 | Unregistered CommenterJoshua

@Joshua

>PMO?

Doh. PMR. Not even close on the keyboard, don't know what I was thinking of.

I think tensions are the mirror view of the way I tend to look at things; i.e. for a diverse population constrained by narrow cultural adherence, I presume the tensions will hit hardest those who happen to benefit least from the dominant consensus narrative (I focus on the latter). So very simplistically, and using your father / mother concepts as flag narratives, there will be more tension for women in a patriarchal society, and more tension for men in a matriarchal society. I think these tensios should just reflect whatever happened to emerge as the main cultural narrative (more usually a set of closely related co-evolved narratives, in practice). And while these are not always so simple as the examples, hence the tensions are likely complex / subtler too, there are certainly some very blunt / simple examples in history; e.g. rather a lot of tension being a Jew in 1930s/40s Germany. You prompt me to wonder whether in some more subtle cases the tensions may be more obvious than that which creates them, meaning they can be used as a detection system; will have to think about that.

Nevertheless I'm not seeing a primary place for "nurturing/disciplining", but perhaps your search will hit gold one day. Meanwhile there are thoughts that chime very much with mine, for instance:

>One example might be "individuality vs. social consensus"

Every strong cultural consensus will generate an opposing reaction; this *is* foundational imo, because it doesn't depend upon what the consensus is about. The reaction is part of the evolutionary rules, and stems from a long evolved arms race between cultural deception and detection, though the detection system itself is also a function of cultural values. How the reaction is framed will depend on what the original consensus is about, and also which allied cultures may get dragged into each side of the clash, but an individuality theme of some kind would certainly fit the bill in some cases. However at my last pass through this effect, you very much disagreed with the principles 0:

I think your 'unknowableness' is another rich seam that looks like gold to me. Exactly where no answer is possible, due to unresolvable uncertainty or (in your bringing up children example) no real possibility of an ideal equation ever for the generic case anyhow, is exactly where cultural narratives thrive; uncertainly is like high octane fuel for emotive memes. And social inertia means that even in cases where the uncertainty can be, and eventually is, much reduced, an exploitative (i.e. via selection, not agential) culture that sprang up before this happened, can remain robust for long long afterwards. Obviously where significant uncertainty remains, the fuel is always high grade; meaning as you say polarization / demonization / etc. Smaller subset issues that might on their own have gotten resolved if they'd stayed below the cultural radar (you mention some), are constantly re-ignited by the larger clash, rather like when superpowers constantly keep small proxy wars stoked in places where their interests overlap.

December 11, 2017 | Unregistered CommenterAndy West

another essential read - genetics explains 12% of delay discounting (present bias) variation:
http://neurosciencenews.com/adhd-obesity-reward-genetics-8160

December 11, 2017 | Unregistered CommenterJonathan

Funny coincidence that I came across this tonight... The relevant part is from around 2-11 minutes. I'm not a big fan of Brett, but what he says there tracks rather closely with some of what I was going for above:

https://youtu.be/rpb-COhbMIM

December 11, 2017 | Unregistered CommenterJoshua

link drop - they saw a shooting:
https://phys.org/news/2017-12-racial-political-identities-people-view.html

This surprises me a bit, considering other recent research showing that political identities are becoming more important than race/class distinctions on a wide range of subjects. Could this difference be due to the Kaepernick effect?

Can't find a non-paywall version, so can't check the survey dates vs. Kaep's actions, which started 9/16.

December 13, 2017 | Unregistered CommenterJonathan

Jonathan -

"Our evidence suggests that racial identities will have the greatest influence."

From the article, I didn't get how they draw that conclusion. They say that black conz are more likely than white conz to see a widespread racial factor. But it isn't clear to me how they get from that to conclude that the racial identity is stronger than the ideological identity. Why couldn't it just be that the racial identity comes into play? Obviously, it seems to me, more black libz would find a widespread problem than black conz.

??

December 13, 2017 | Unregistered CommenterJoshua

Also, I don't get why you think that Kaep would have a significant influence:

(1) I'm sue that most blacks have seen a racial influence in their treatment at thanks of the polices for decades prior to Kaep.

(2). If anything, I would imagine that the Kaepernick effect would increase the relative influence of ideology compared to race than the other way around. Black conz see more reason to dig deeper into their foxhole because of Trump enhancing the ideological divide over Kaepernick. Or is that what you WERE suggesting (as I reread your comment I can't tell if you might be saying that the study was conducted pre-Kaepernick, and thus might not reflect the more recent trend?

December 13, 2017 | Unregistered CommenterJoshua

Joshua,

"They say that black conz are more likely than white conz to see a widespread racial factor. But it isn't clear to me how they get from that to conclude that the racial identity is stronger than the ideological identity."

True. This is why I hate to rely on popsci sometimes (OK, most-times). By saying "Racial identity appeared (as the) most powerful predictor of perceiving police force as a widespread problem, the researchers found, even trumping ideological identity." they could mean that the correlation between race and opinion was higher than between political identity and opinion. Without the actual study, though, this is just my guess. I have seen cases where phys.org got things wrong - summarizing in plain English a conclusion that a study doesn't find, even getting it backwards.

As for Kaep, my point is just to question how things may have changed because of the NFL anthem protests that Kaep initiated. In other words, does the timing of the study vs. those protests make the protests relevant or not to the study's conclusions.

December 13, 2017 | Unregistered CommenterJonathan

If some self-sorting process is going on in western societies that either pushes high S2'ers somewhat left or enhances S2 somewhat when on the left, then why doesn't this happen to MS2R or SC?

That's a basic question, Jonathan - basic and difficult in the most interesting way.

My hypothesis: MS2R and SC are both closer measures of cognition than S2, and S2 activity is actually a worldview. I've long suspected that S2 is only considered "cognitive" because of the dualist history of Western thought about cognition that opposes ratio to sensus, "thoughts" to "feelings".

December 25, 2017 | Unregistered Commenterdypoon

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>