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

by on December 1, 2012 at 7:31 am in Education, Philosophy, Political Science, Religion, Science | Permalink

That is a new paper by Dan M. Kahan, at Yale Law School, and it has to do with what I call “mood affiliation”:

Social psychologists have identified various plausible sources of ideological polarization over climate change, gun violence, national security, and like societal risks. This paper reports a study of three of them: the predominance of heuristic-driven information processing by members of the public; ideologically motivated cognition; and personality-trait correlates of political conservativism. The results of the study suggest reason to doubt two common surmises about how these dynamics interact. First, the study presents both observational and experimental data inconsistent with the hypothesis that political conservatism is distinctively associated with closed-mindedness: conservatives did no better or worse than liberals on an objective measure of cognitive reflection; and more importantly, both demonstrated the same unconscious tendency to fit assessments of empirical evidence to their ideological predispositions. Second, the study suggests that this form of bias is not a consequence of overreliance on heuristic or intuitive forms of reasoning; on the contrary, subjects who scored highest in cognitive reflection were the most likely to display ideologically motivated cognition. These findings corroborated the hypotheses of a third theory, which identifies motivated cognition as a form of information processing that rationally promotes individuals’ interests in forming and maintaining beliefs that signify their loyalty to important affinity groups. The paper discusses the normative significance of these findings, including the need to develop science communication strategies that shield policy-relevant facts from the influences that turn them into divisive symbols of identity.

To put that in Cowenspeak, both sides are guilty, the smart are guiltiest of them all, and the desire for group loyalty is partially at fault.  Is it possible you have seen these propensities in the economics blogosphere?

Here is a related blog post by Kahan, here is another on how independents do somewhat better than you might think, here is Kahan’s blog.

I would stress the distinction between epistemic process and being more right about the issues at a given point in time.  Even if various groups of individuals are epistemically similar in terms of how they process information, at some point in time some groups still will be more right than others, just as some sports fans, every now and then, are indeed backing the winning teams.  It’s less a sign of virtue than you might think.

For the pointer I thank Jonas Kathage.

1 Anon. December 1, 2012 at 7:42 am

Well, that’s depressing.

2 Claudia December 1, 2012 at 8:41 am

Nah. I find the comment section depressing (see the last post as an example)…I couldn’t understand how smart people can stubbornly hold such strong (and yet partially wrong) views. Self included, of course. At least this paper suggests it’s tied to (group) identity…and not emotion or intellect.

I do think this would be a great time to jettison the fuzzy and now we see misleading label of ‘mood affiliation bias.’ It’s not about emotions. How about ‘identity affiliation bias’ or ‘group affiliation bias?’

3 Greg G December 1, 2012 at 9:13 am

Well said Claudia. No doubt Tyler is talking about something real and important here but “mood affiliation” does a terrible job of naming it.

4 TA December 1, 2012 at 9:18 am


5 Carolus December 1, 2012 at 9:18 am

Stuff the mumbo jumbo. Call it what it is — group think.

6 Slocum December 1, 2012 at 10:19 am

Agreed. I think ‘Group Affiliation Bias’ is more apt. And that’s the term that Arnold Kling used when referring to Kahan’s work earlier this year:

7 efp December 3, 2012 at 1:00 pm

Shorter: tribalism.

This is distinct from “mood affiliation,” but probably the main source of it.

8 derek December 1, 2012 at 11:10 am

It is quite simply about power. All the four spheres mentioned in the abstract have attached to them vast sums of money and vast amounts of real power over millions of people and either the economic or personal activity of all those people. Or whether they exist or not. The well educated have always displayed a tendency towards authoritarianism, if only I had the power to implement what I, the smartest in the room, know.

And as usual, those who would get in the way are anything from simply wrong to having something wrong with them to we would be better off without them.

Yes, power corrupts, absolute power corrupts absolutely. It is depressing.

9 Willitts December 1, 2012 at 2:56 pm

I think you’re missing an important aspect of public choice here.

Suppose I oppose/support gun control to some substantial degree. However, I am either A) willing to accept some compromises to form a solid, enduring consensus or B) more moderate than most of the people on my side on SOME aspects/dimensions of that broad policy.

Even if there is some optimal policy for the population along every dimension of “gun control,” we are unlikely to find it because there is and cannot be a perfect system of preference revelation. And, quite possibly, there are multiple equilibria that have distributional impacts on individual utility; individuals will therefore apply power and influence in order to shift the equilibrium.

So we have an existence problem, a uniqueness problem, and a how do we get there problem.

Compound that with the fact that preferences can change (in an honest debate), enforcement of the optimal agreement could be abused in favor of one side, the issue may be peripherally related to other issues, and there are reputational effects both within and between factions in a dynamic competitive political system, hence….

I don’t believe all people are rational nor do I believe voting systems work nor do I think there are no cognitive biases. But the situation described above can exist without any of these problems. In other words, the game we’re playing is far more complex than any of you give it credit for.

And what I see developing here is the notion that people with strong preferences are “stubborn” or “extreme.” As Barry Goldwater said, “extremism in the defense of liberty is no vice.” An equally powerful statement can be made for the defense of other important values.

And to put a finer point on it (for those who call it ‘group think’), people may be willing to compromise with an apparent winning coalition merely to put the argument to rest. In other words, people accept less than optimal outcomes when there are costs to either the status quo, continued confrontation, or the possibility of a worse outcome. This is not irrational and it is not dispositive of our notion of ‘group think.’

An extreme example of a dysfunctional system of choice is the so-called Abilene Paradox where a group of people voluntarily agree on a choice that is strictly dominated by another available choice. The paradox usually involves A) imperfect preference revelation; and B) interdependent utility functions.

10 Brian Donohue December 2, 2012 at 8:56 am

Excellent comment.

11 Michael G. Heller December 1, 2012 at 7:57 am

As an openly ideological person I liked this bit best: “subjects who scored highest in cognitive reflection were the most likely to display ideologically motivated cognition”. As a person who — in order to maintain ideological and Cartesian ideological purity — couldn’t give a bugger about loyalty or what others think of me I disliked this bit most: “findings corroborated [blah theory that] information processing that rationally promotes individuals’ interests in forming and maintaining beliefs that signify their loyalty to important affinity groups”.

You know things get really complicated when 1. you assume groups have thoughts right or wrong (only individuals think, and to say — based on this sort of ‘evidence’ — that they are right or wrong is an ideological statement) or 2. you consider the problem of ideological polarisation over the issue of ideological polarisation.

12 Rahul December 1, 2012 at 9:34 am

What does “Cartesian ideological purity” mean?

13 Lord December 1, 2012 at 10:41 am

I am not so sure this is about group affiliation so much as group definition. Groups don’t exist in abstract; they are led and have followers, so these are less likely to be following groups than leading them.

14 Dredd December 1, 2012 at 8:51 am

A noted evangelist and a quantum physics professor both renounced some of their long-held positions on the age of the Earth and supersymmetry this week.

Childruns do learn.”

15 Brian Donohue December 1, 2012 at 10:20 am

As an Independent, I’m cautious about too readily accepting the link that suggests we’re smarter than y’all- self-serving confirmation bias and such.

Food for thought though.

16 lxm December 1, 2012 at 10:21 am

” Is it possible you have seen these propensities in the economics blogosphere?”

No, not ever.

17 Dismalist December 1, 2012 at 11:12 am

But this shouldn’t be surprising: Ideology is nothing more than cheap theory. It enables one to jump to conclusions without doing too much work. Smarter individuals have lower information processing costs, so its easier for them to make news fit their ideology. Thus, ideology, is subject to a Darwinian struggle. Takes a while to declare a [temporary] victor, but man, the victories tend to be overwhelming. Viz. 1989.

18 gorobei December 1, 2012 at 11:14 am

Junk paper. We eventually see graphs on page 19, and they’re based on Monte Carlo simulations, and are plotting discrete data as if it were continuous. The graphs we encounter later are even worse.

Nothing to see here.

19 Gordon Mohr December 1, 2012 at 12:32 pm

Seems like more support for the “politics isn’t about policy” view. (It’s about signalling group loyalties and which people/types-of-people should have status.)

I wonder, is this effect timeless, or has it been getting worse over time?

My conjecture: it’s getting worse. The political class is larger: We have 24/7/365 cable- and internet- politi-tainment, making it easier to indulge in political loyalties like sports team fandom. (The mapping of parties to ‘red’/’blue’ colors, which used to vary between elections, has hardened.) People in general can more afford the luxury of armchair politics (outside of occasional, brief election seasons)… and more casual participants means more participation at the identity/signalling level (even if those same people are more coldly rational, by necessity, in their ‘day jobs’).

20 msgkings December 1, 2012 at 2:09 pm

Nice post, I agree. Sadly.

21 derek December 1, 2012 at 2:32 pm

I would disagree. There is a myth of some time in the past where everyone got along. It was more a description of a time where one opinion had ascendancy due to some unusual series of events.
Consider the differences in opinion as to what to do about the current economic malaise. One side wants to expand the money supply and the reach of government in an industrial planning sense. Otherwise it is the end of the world. The other side says no, shrink government, tone down the Fed or else we will see the end of the world. Real differences in opinion each with some data points in support.
An anecdote. An elevator was being installed where I was working. Chatting with the installer he said the boundaries between his trade and the electricians were established in street fights in New York.
All these disagreements have real consequences for real people. In January lots of people are going to have less money than they did in December. Should everyone just submit to the divine will of our fearless leaders?

22 Willitts December 1, 2012 at 3:06 pm

I agree with everything you said, but I don’t think what you said and what Gordon said are mutually exclusive.

23 Bernard Guerrero December 3, 2012 at 12:21 pm

I believe it is largely timeless. Note that the period of broad political consensus (such as it was) which allowed for the appearance of a Cronkite-esque neutrality was a post-WW2 phenomenon. As has been noted previously in this blog, the norm has actually been a cheap, no-holds-barred sort of partisanship which stretches back to papers like Callender’s Aurora publishing any nasty bit of stuff they could find or make up about Hamilton and other Federalists, and other organs doing the same with regards to Jefferson and co. For that matter, you can trace the sports-like associations all the way back to the Greens and Blues.

My supposition is that the interlude was due to two factors: A) the external enemy, international Communism, was unpleasant enough for all factions to be willing to subsume their pettier disagreements within a larger “unity” narrative, and B) the advent of radio and then TV required a scale that was available only by going after a very large audience, which meant that the common elements of conventional wisdom were emphasized at the expense of stuff that might alienate part of your audience. The Net has to some extent weakened the need for the latter while the death of Communism has removed the former, allowing the natural pattern of in-group/out-group and news that confirms your preexisting beliefs to return.

24 Urso December 3, 2012 at 4:41 pm

Post-WWII American politics was no mutual admiration society. What of the Civil Rights struggle? I don’t see riot police getting called out in the streets today with the frequency that they were in the 60s; even the OWS protests were quite harmless in comparison.

I think this gets papered over today, partly because we (we being “white people”) would all rather pretend that all right thinking Americans were in favor of the CRA, even at the time, and that it was just a few rural Southern reprobates holding racial equality back. Hardly.

25 Anthony December 5, 2012 at 2:43 pm

The Civil Rights struggle was the biggest issue of the late ’40s through the mid ’60s, but was almost entirely an internecine struggle among Democrats, and was largely a region vs region conflict even there, so the big conflict was contained in a way where the vitriol didn’t spill out into public as much.

26 Douglas Knight December 1, 2012 at 1:54 pm

Cognitive reflection isn’t the same as intelligence. I would expect intelligence to make things worse and cognitive reflection to make things better. Since they are correlated, that is not inconsistent with this result, but it’s bad in any event.

27 Joe Smith December 1, 2012 at 2:18 pm

So smart people can talk themselves into believing stupid things – hardly new. Neither is it new that a little knowledge can be a dangerous thing.

28 Josh Broder December 1, 2012 at 2:45 pm

Where does the individual libertarians with extremely high cognitive reflection (and from the pre-internet age) who resisted the group think & the group loyalty of the left-center cult dominating the universities fit in here? They would seem to be signal counter-examples to the conclusion of this paper.

29 Ryan December 1, 2012 at 5:55 pm

Such a person fits right in as soon as said person refers to themselves as a libertarian. They are an example, not a counter example, of the conclusions of the paper.

30 Joe Smith December 1, 2012 at 5:58 pm

No. Libertarians as a group may be the ultimate expression of this process at work – they are just adhering to their own special group. They may have been initially triggered by alienation or social dysfunction but once they latch on to Libertarianism as a world view that legitimizes their alienation from their peers they are hooked and can rationalize the rest.

31 Brian Donohue December 2, 2012 at 9:12 am

It may seem that way to you Joe, but c’mon, libertarians a are a scattered little band, made up of disagreeable people.

A comment section like this board is about the only place you can find a ‘libertarian community’, and it’s still non-stop bickering once the subject strays from the narrow but (in libertarians views) crucial and far-reaching one of economic policy.

But the mere existence of this board impels liberals, not content with hegemonic control of vast organs of power, to come over here and kick the furniture.

Which I, and most libertarians, welcome. My feeling is, if you don’t do everything in your power to tear your own arguments down, someone that you probably find unpleasant will do you that favor.

32 Ryan December 1, 2012 at 5:50 pm

AH! The postmodernists were right! NOO!

33 Sam December 1, 2012 at 9:29 pm

Very interesting links, Tyler. Especially (disturbingly ironical) interesting to read the reasonings.
A terrific demonstration of Group Think/Bias is on display in the Google video where Hal Varian hosts Nate Silver for a discussion of the election results; what is wonderfully ironic about it is that Nate was dead right, but there is a self-congratulatory tone to the event (though Nate was astonishingly subdued in this regard). And this bias was reflected in the rightwing blogosphere – as Nate notes, too, of Romney supporters either heartened by signs in the yard in their neighborhoods, or misleading themselves as to the likely outcome of the voting. And the Libertarians did not fully support their best candidate in some time, Gary Johnson (except, perhaps, in DC!?)

34 Sam December 1, 2012 at 10:56 pm
35 y81 December 1, 2012 at 11:05 pm

Only among rather small, boring, self-satisfied groups like the Yale faculty (I spent four weary years doing field research, so to speak) is the surmise common that conservatism is distinctly associated with closed-mindedness. Most of my friends think that there is a rather more distinct association between university professorship and laughable insularity. But hey, we give them money, they give us diplomas, live and let live.

36 TGGP December 1, 2012 at 11:37 pm

My co-blogger found some research indicating that using a smaller font can reduce the partisan filtering of information.

37 eddie December 2, 2012 at 12:46 pm

“ideological polarization over gun violence”

There is no polarization over gun violence. Everyone is against it.

What we’re polarized about is ‘gun control’, which itself is a euphemism for government restrictions on gun ownership.

38 TGGP December 5, 2012 at 1:52 am

A charitable reading would be that the polarization is not over whether it is good or bad, but what causes it and how we should respond.

39 J1 December 2, 2012 at 7:54 pm

“the hypothesis that political conservatism is distinctively associated with closed-mindedness”

Talk about closed-mindedness. At least the authors have the intellectual integrity to admit this is baloney.

40 Steve Roth December 4, 2012 at 9:25 pm

Worth looking deeper on this one. Here’s the three-question “cognitive bias” test they used:

1. A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball.
How much does the ball cost?
2. If it takes 5 machines 5 minutes to make 5 widgets, how long would it take
100 machines to make 100 widgets?
3. In a lake, there is a patch of lily pads. Every day, the patch doubles in size.
If it takes 48 days for the patch to cover the entire lake, how long would it
take for the patch to cover half of the lake?

They assert that “CRT is performance-based measure of subjects’ disposition to engage intellectual problems in a reflective and open-minded manner, and it has been shown to be superior to self-evaluation measures in predicting vulnerability to the various cognitive biases associated with low-level information processing”

But it is in fact just a measure of the willingness to “think twice” (level 2 thinking) in numerical problem solving.

As for the studies they cite supposedly correlating the test with “various cognitive biases,” well I’ll just share a couple of those correlations with you from one of those studies:

Thinking Dispositions
12. Actively Open-Minded Thinking .10
13. Superstitious Thinking .04

So much for being a measure of “open mindedness.”

Their Dem/Pub comparison chart seems, then, to mainly compare willingness to think hard about a numerical problem. In that study’s words, culminating the abstract:

“CRT is a particularly potent measure of the tendency toward miserly processing”

Cognitively biased people can spend exceptional amounts of brain energy maintaining those biases.

Is it possible you have seen these propensities in the economics blogosphere?

41 Steve Roth December 5, 2012 at 11:24 am

Exploring more, this is what the study appears to show:

When a given test that does not on its face appear to be a valid predictor of self-serving biases (ideologically motivated thinking, open-mindedness):*

Democrats and Republicans are equally likely to question the validity of that test.**

They are equally likely to change that belief based on their self-serving biases.

I would ask:

How would those results differ if the test did appear to be valid on its face?

This study could have used a test of political or scientific knowledge as its benchmark instead, I’m guessing with similar results. An equal number might consider such a test to be a valid (invalid) measure of open-mindedness or cognitive bias.

* And in fact is not a valid predictor of those traits, though that is immaterial to this part of the results.

** When asked to evaluate “I think the word-problem test I just took [i.e., the CRT test] supplies good evidence of how reflective and open-minded someone is,” both groups cluster (symmetrically) around “slightly disagree.“

42 John Lilburne December 5, 2012 at 11:50 pm

The Boxer- Simon and Garfunkel

I am just a poor boy
Though my story’s seldom told
I have squandered my resistance
For a pocket full of mumbles such are promises
All lies and jests
Still a man hears what he wants to hear
And disregards the rest

When I left my home and my family
I was no more than a boy
In the company of strangers
In the quiet of the railway station running scared
Laying low, seeking out the poorer quarters
Where the ragged people go
Looking for the places only they would know

Lie la lie …
[ Lyrics from: ]
Asking only workman’s wages
I come looking for a job
But I get no offers,
Just a come-on from the whores on Seventh Avenue
I do declare, there were times when I was so lonesome
I took some comfort there

Lie la lie …

Then I’m laying out my winter clothes
And wishing I was gone
Going home
Where the New York City winters aren’t bleeding me
Bleeding me, going home

In the clearing stands a boxer
And a fighter by his trade
And he carries the reminders
Of ev’ry glove that layed him down
Or cut him till he cried out
In his anger and his shame
“I am leaving, I am leaving”
But the fighter still remains

Lie la lie …

Comments on this entry are closed.

Previous post:

Next post: