Article on confirmation bias for the Smith Alumnae Quarterly

post by James_Miller · 2014-08-06T14:43:11.412Z · LW · GW · Legacy · 24 comments

Contents

24 comments

The head of the IMF was supposed to be Smith College's commencement speaker, but withdrew because of faculty and student protests.  A few professors (although none in the economics department) wrote to the IMF chief asking her to cancel.  The Smith Alumnae Quarterly asked several people, including myself, to write a 400 word article on the surrounding issues of diversity of thought and protest.  Below is a draft of my article.  I hope it's of interest and I would be grateful for any suggestions for improvement:

When you’re looking for a guest speaker for an important event exclude anyone who advocates sacrificing children to the demon Moloch.  I’m all for freedom of speech, but poisonous crazy weeds can destroy an intellectual garden.  Unfortunately, because of confirmation bias, when we encounter an objectively reasonable political belief that we disagree with, that belief often feels dangerous and absurd. 

Humans don’t naturally search for truth; rather we seek to confirm our pre-existing beliefs.  We tend to remember evidence supporting our positions, but conveniently forget conflicting data.  Consequently, the opinions of political opponents can falsely seem ridiculous because they cut against all of the relevant facts stored in our brains.  

Closed intellectual ecosystems containing homogenous political beliefs, such as the Smith College bubble, provide fertile grounds for confirmation bias.  When almost everyone shares the same political opinions, nonconformists will tend to shut up (or transfer) to avoid paying the social cost of dissent.  The majority will therefore never encounter people who challenge their beliefs and so most people can comfortably categorize political opponents as dangerous others. 

Having a high IQ can also worsen confirmation bias.  Being smart makes it easy to rationalize away uncomfortable truths.  And many of us among the academically gifted derive a huge amount of self-worth from thinking that WE ARE RIGHT.

You know that confirmation bias corrupts your thinking if you dismiss the opinions of experts on topics you don’t really understand.  For example, pretend that although you know a little bit about economics from watching MSNBC and reading New York Times editorials, you have never done the graduate level work in mathematics and statistics necessary to properly evaluate what the IMF does.  If you still decided that the head of the IMF is unfit to speak with the people you are paid a lot of money to educate, then confirmation bias comfortably feeds on your brain.

Fight confirmation bias by asking of your most cherished beliefs, “What evidence would convince me that I’m wrong?”  Here is an exercise for readers devoted to social justice:   Feminists demonized then-Harvard president Larry Summers for suggesting that genetics might partly explain the relative lack of women in science.  We are going to learn a lot more about the genetic basis of intelligence over the next decade.  What evidence would convince you that Summers was correct?  (Summers, of course, went on to apologize profusely for his heresy, doing everything short of sacrificing his own children to show repentance.)  

24 comments

Comments sorted by top scores.

comment by Varuna · 2014-08-06T21:57:50.321Z · LW(p) · GW(p)

Your use of the Summers incident seems far more likely to inflame people than help educate. I predict it will increase the number of readers whose takeaway from the article is that you simply want to defend the indefensible, and are best to be ignored. People who already agree with you will enjoy it.

comment by fubarobfusco · 2014-08-07T08:33:46.782Z · LW(p) · GW(p)

This is not an article about confirmation bias. It is an article about your political beliefs, rationalized by accusing your opponents of confirmation bias. It does both subjects a disservice.

Replies from: James_Miller
comment by James_Miller · 2014-08-07T14:10:25.812Z · LW(p) · GW(p)

If you have the time, I would be grateful if you provided more of a justification of this, but I will understand if you don't however as written your criticism doesn't provide any guidance as to how I can better write the article. My main political identity is being a free market economist, and like many of my type I do not support the IMF.

Replies from: fubarobfusco
comment by fubarobfusco · 2014-08-07T23:47:30.202Z · LW(p) · GW(p)

If you are actually interested in writing an article about confirmation bias, use examples that are not political flamebait to the community you're speaking to. Doing so primes your audience to dismiss you as an axe-grinder or troll, and thus to mistakenly associate the idea "confirmation bias" with hostility ... or just with your particular political position. Cognitive biases are bigger than your political position; don't diminish the science by implying that rejecting your politics implies rejecting the science.

If you are interested in making political points about Smith College not being welcoming toward Christine Lagarde, do not present it as an article about confirmation bias. Doing so is intellectually dishonest. Instead, investigate and respond to the arguments made by those who objected to Lagarde's invitation. To their argumentsnot to the psychological processes you conjecture are behind them.

It is pretty much always poor form to psychoanalyze your political opponents and present their beliefs or behaviors as a consequence of the pathology you ascribe to them. Doing so is a failure to leave a line of retreat, and is is also a form of the genetic fallacy — even if you're right about the pathology, just because I'm crazy doesn't mean I'm wrong.

Replies from: James_Miller
comment by James_Miller · 2014-08-08T01:25:47.799Z · LW(p) · GW(p)

Thanks, you make some good points. Reading your comments caused me to realize that I'm not interested in taking the time to find out why the professors didn't want Lagarde to speak at Smith because I assign a low probability to my finding their arguments reasonable. (The time I would need to spend doing this could be much better used, for example, reading your past LW contributions.) I don't think this is because of confirmation bias, but of course if it were I wouldn't think it was.

The first sentence was supposed to be a line of retreat in which I admitted that it is appropriate to exclude some people.

It is pretty much always poor form to psychoanalyze your political opponents and present their beliefs or behaviors as a consequence of the pathology you ascribe to them

Poor form perhaps, but not necessarily inaccurate.

Replies from: fubarobfusco
comment by fubarobfusco · 2014-08-08T21:54:24.033Z · LW(p) · GW(p)

I'm not interested in taking the time to find out why the professors didn't want Lagarde to speak at Smith because I assign a low probability to my finding their arguments reasonable.

I expect your opponents think the same of you; albeit with different phrasing. And thus by symmetry you each defect against the other, and thus is elucidated the old theorem regarding the bitterness of academic disputes.

Replies from: James_Miller
comment by James_Miller · 2014-08-08T22:18:25.479Z · LW(p) · GW(p)

Agreed!

comment by Lumifer · 2014-08-07T15:10:14.210Z · LW(p) · GW(p)

From my perspective you're glossing over the distinction between rationality and values and that opens multiple ways for dismissing your arguments. For example,

You know that confirmation bias corrupts your thinking if you dismiss the opinions of experts on topics you don’t really understand.

Not necessarily -- if I know that the opinion of experts is driven by certain values and I reject these values, I can reasonably dismiss the experts' opinions even if I don't understand the technicalities. In crude terms, I don't care how they propose to get to X if I don't want to go to X.

Same thing:

...you have never done the graduate level work in mathematics and statistics necessary to properly evaluate what the IMF does.

A lot of people in your audience reject "what the IMF does" on the basis of values. Grad level math is entirely unnecessary for that.

You kinda recognize that in the beginning by admitting that you don't want to invite Moloch worshipers even if you don't have a graduate degree in religious studies.

One more nitpick/question -- what in the world is an "objectively reasonable political belief" and how do you distinguish it from an "objectively unreasonable" one? Do you just mean a correct view of reality?

Replies from: James_Miller, Viliam_Bur
comment by James_Miller · 2014-08-07T15:16:37.591Z · LW(p) · GW(p)

Thanks for the constructive criticism.

Do you just mean a correct view of reality?

Yes, but I didn't want to mention the T word. There is only so much inferential distance you can cover in 400 words.

Replies from: Lumifer
comment by Lumifer · 2014-08-07T15:36:31.403Z · LW(p) · GW(p)

I didn't want to mention the T word

A wise choice, probably :-) but you can also phrase it something like "consistent with what we see in reality"...

comment by Viliam_Bur · 2014-08-08T13:12:59.273Z · LW(p) · GW(p)

A lot of people in your audience reject "what the IMF does" on the basis of values.

Research shows that people with certain political opinions are much worse than average in understanding what their opponents believe. A lot of "rejecting on the basis of values" is probably based on this.

(Step 1: I completely strawman my opponent, including his values. Step 2: I "rationally" reject my opponent's conclusion, despite his arguments and expertise, because it was based on such horrible values.)

Of course, this is another inferential step, too much for one article. Also, the argument applies recursively: why should anyone trust the research, if it was probably done by people with evil values?

Replies from: Lumifer
comment by Lumifer · 2014-08-08T15:18:37.902Z · LW(p) · GW(p)

Research shows that people with certain political opinions are much worse than average in understanding what their opponents believe.

I tend to quite sceptical of research which claims to show that people from the enemy political tribe make up stuff all the time and are just stupid :-/

A lot of "rejecting on the basis of values" is probably based on this.

I'm not too sure about that. Rejection on the basis of values is a very simple and basic operation. Consider e.g. abortion or gun debates -- the issue is not that one sides misunderstands reality.

Replies from: Viliam_Bur, VAuroch
comment by Viliam_Bur · 2014-08-09T21:37:49.525Z · LW(p) · GW(p)

I imagine gun debates like this:

A: I am against guns, because guns kill people and I am against killing people. I am horrified that someone likes killing people.

B: No, I don't like killing people, that's nonsense. I believe that having guns would reduce killings. Here is my model that kinda supports my argument...

A: Nice try! But I know you truly are just another nutjob who likes killing people. See, we have fundamentally different values, that's why we can never agree.

Okay; I admit that I am not an American, and I have never debated any Americans on this topic. I am just extrapolating from my general model of people. But I have a feeling that if I went to a random progressive American university, chose dozen random anti-gun students and asked them: "explain to me why some people are pro-guns", most of them wouldn't give an answer "because according to their model (which is factually wrong, of course), owning guns prevents human deaths". Instead I would probably hear something about crazyness or religion or something like that. (If there actually was some research about this, I would like to see the results.)

Replies from: IlyaShpitser
comment by IlyaShpitser · 2014-08-09T22:19:04.355Z · LW(p) · GW(p)

You can't drink alcohol on a beach in New Jersey (without a permit). Why? Because some idiot was unreasonable once, and someone complained, and now there is a universal quantifier in place.

It's hard to draw lines against scoundrel behavior in a way that avoids undesirable side effects for reasonable people (law is hard because it is a kind of applied analytic philosophy).

comment by VAuroch · 2014-08-09T09:03:17.563Z · LW(p) · GW(p)

Maybe they don't misunderstand reality, but they may well misunderstand what the other side means when they state their objections.

Replies from: Lumifer
comment by Lumifer · 2014-08-10T00:09:04.203Z · LW(p) · GW(p)

The abortion and gun control debates are loud and public. Do you have a specific misunderstanding to point to?

Replies from: VAuroch
comment by VAuroch · 2014-08-10T00:51:58.991Z · LW(p) · GW(p)

For those two, sure. What 'pro-life' means to people who consider themselves pro-life, and what people who consider themselves pro-choice model it meaning, are not the same. Similarly, gun control people largely have a very different relationship with and understanding of guns, and so don't have an accurate conception of what gun rights people see themselves as protecting. In both cases, the reverse is also true.

comment by Shmi (shminux) · 2014-08-08T21:37:59.109Z · LW(p) · GW(p)

Recall that the original meaning of "politics is the mind killer" was not an Eliezer's injunction against mixing politics and rationality, but a warning against unnecessarily using political examples to make a non-political argument:

What on Earth was the point of choosing this as an example? To rouse the political emotions of the readers and distract them from the main question?

Your examples of IQ, MSNBC, IMF and feminism are unnecessarily polarizing if your goal is to discuss confirmation bias.

comment by KnaveOfAllTrades · 2014-08-06T23:47:11.136Z · LW(p) · GW(p)

If you still decided that the head of the IMF is unfit to speak with the people you are paid a lot of money to educate, then confirmation bias comfortably feeds on your brain.

I'm not sure how strongly that holds, but it's worth mentally noting that "The most qualified people to make policy decision D are all F's, so anybody who is not an F should defer to any F on D" is a logical fallacy, albeit possibly strong evidence, depending on context.

For example, maybe all the most qualified experts on the existence of God(s) are philosophers. But given LW's broad atheism, I reckon the people who upvoted your post would have a hard time accepting that a weekend atheist should defer to a theistic philosopher. Someone could start out with the basic beliefs of the head of the IMF and be polarized further in those beliefs by their studies of economics, only remembering the economic evidence favouring their position. If one thinks this happens about equally to those one agrees with and those one disagrees with, then one's posterior ends up being the same as the prior; one expects to see lots of disagreeing experts regardless of whether one is correct.

Replies from: James_Miller
comment by James_Miller · 2014-08-07T00:10:20.860Z · LW(p) · GW(p)

You need to be far more certain of your beliefs to think that (1) "this F's views are so evil that she should be excluded from speaking to our community, compared to (2) "I shouldn't defer to this F's position on D."

I do agree with you that it's dangerous to say "my field dominates the study of this topic and you are not in my field so defer to the judgement of my field when looking at this topic" because some fields have not developed techniques which actually help them ascertain truth.

Replies from: KnaveOfAllTrades
comment by KnaveOfAllTrades · 2014-08-07T00:19:47.801Z · LW(p) · GW(p)

Good point!

comment by IlyaShpitser · 2014-08-08T20:20:13.722Z · LW(p) · GW(p)

Too many fnords in this.

comment by blake8086 · 2014-08-06T17:48:04.429Z · LW(p) · GW(p)

I really enjoyed that.

Instead of simply saying "fight confirmation bias", try to give reasons why to fight confirmation bias. Being less wrong is often not rewarding enough for people.

Fighting confirmation bias makes you sexy! It will make your peers all think you're smart! You'll feel better about yourself!

You say "many of us among the academically gifted derive a huge amount of self-worth from thinking that WE ARE RIGHT." How could you redirect some of that perceived potential self-worth gain into fighting confirmation bias?

Replies from: James_Miller
comment by James_Miller · 2014-08-06T17:49:26.313Z · LW(p) · GW(p)

How could you redirect some of that perceived potential self-worth gain into fighting confirmation bias?

Excellent idea!