Thursday, September 16, 2010

Cognitive Biases Are Cool

I recommend to everyone excellent posts on cognitive biases, misinformation, and partisanship by Brendan Nyhan, who is interested in the question of why people continue to hold false beliefs about Barack Obama, and John Sides, who summarizes some fascinating research. While I'm interested in both cognitive biases and partisanship, I'm far from an expert in this area, so I'll leave it to Brendan and John to feed you the literature, and stick to one quick anecdote.  My favorite experience with this was in fall 2000, which happened to be my first time teaching a very large section of an introductory course in American Politics.  Of course, we did discuss in class the extraordinary events surrounding the Florida recount.  It rapidly turned out that not only did all my students turn out (shockingly!) to have deeply felt beliefs about whether hand or machine methods were the correct way to count (and/or recount) ballots, but they had, or so they reported, held those beliefs for some time.  Oh, and (and all this was by show of hands, class discussion, and plenty of after-class discussion) by some amazing coincidence their long-held beliefs about the superiority of human or machine counting lined up almost exactly with their support for either George W. Bush or Al Gore.  What I really like about this one was that unlike most policy positions, this one was transparently generated by the partisan situation; of course, in reality none of my students had ever thought for a moment about the best way to count ballots until Election Day 2000, and there is no logical "liberal" or "conservative" position on how to count ballots (that's true of many issues, but was painfully obvious in this particular case).  That was not, however, what it seemed like to them at the time. 

And part of what makes the topic so fun, at least in my view, is that there's also a bias towards fitting our memories to match our current logic, so I suspect that if I went back to those same students today, ten years later, few of them would even remember passionately holding those positions.  Just as many Democrats over the last five years have gone from relative indifference between Barack Obama and Hillary Clinton (and, don't forget, John Edwards), to intense belief that one was The Forces of Good and the other that season's Big Bad, and now back to the original (more or less) feelings about them, all without realizing any inconsistency or, to a large extent, any active change.

Great topic, with lots of fascinating implications for our theories of politics and democracy.


  1. One of the most humbling experiences of my life came when discussing the election of 1996 a few years ago. I ended up voting for Bob Dole because I had grown weary of Bill Clinton's personal moral failings, and the Lewinsky scandal was the last straw. (For the record, it will be a cold day in hell before I vote for another Republican, but that's a different comment thread.)

    Except the problem is that the Lewinsky scandal broke well after the 1996 election, as was pointed out to me during the conversation. And I was just completely flabbergasted, because I was sure it was the scandal that led me to vote for Dole. For the life of me now, I can't remember what it must have been that actually did influence my vote at the time.

    [This can also be interpreted as a cautionary tale about inhaling.]

  2. Do you think people can overcome cognitive biases? If so, how and how often?

  3. Dan, great story. David...I know there's work on that, but as I said it's not my area of expertise. I think the general answer comes down to yes, in specific cases in which things line up right or with training. Perhaps those who know more than I do will chime in here.

  4. Here's what might have been happening with Dan: by 1996, Clinton had experienced not one but two high-profile "bimbo eruptions". First La Flowers in the 1992 NH primary period, then (iirc) in 1994-5 came the Paula Jones tabloid trash allegations ("Troopergate" and the American Spectator story) and her sexual hayrassment lawsuit from that period.

    Finally, in the 1996 campaign you had Bob Dole loudly lamenting on the stump: "Where's the outrage?" as he seemed to be referencing some of Bill's philandering (along with some other pseudoscandals) as he watched people largely yawn about it in response.

  5. My $.02: perhaps what we label as cognitive biases is merely a special case of what an earlier generation referred to as cognitive dissonance. You recall cognitive dissonance: someone makes a compelling argument that you viscerally dislike, but whose veracity you recognize, though your emotion (visceral dislike, in this case) overcomes your rationality and you come to perceive the argument as false (which mollifyies your hostility, but not your intellect).

    Perhaps this may explain why the educated show greater transition toward irrational beliefs: perhaps educated folks fight more against the dominant (emotional) side of cognitive dissonance (not that they try to, they just may process more information, which tends to attack their emotional biases more often). When the emotions win, as they seemingly always do, there may be a greater whipsaw effect in the educated, who have presumably been fighting said emotions more than the uneducated.

    Finally, 2000: the particulars of the recount matter, of course, at least if you believe the Miami Herald's claim that a full Florida recount would have netted Gore a 3 vote victory in the state. But as far as tragedies go, the real abomination was the idiots who thought the butterfly ballot was a good idea, resulting in an inexplicable 1700 or so elderly Jews voting for Buchanan when their demographic screamed Gore.

    Nothing much could ever be done about the butterfly ballot, so its not a good source of hate. We sure do like to hate.

  6. Kind of like all the "progressives" who were very anti-war until Obama decided to double down on every Bush foreign policy...

  7. That's really interesting. I wonder if it has any correlation to another phenomenon I've wondered about, which is why do people tend to adopt most, if not all, of the views of a political party even where the independent views are not necessarily related?

    Example: The answer to the question whether or not humans have caused global warming is an objective scientific fact. The correct answer has nothing to do with the question whether single payer health care is a sound policy. But most people who are opposed to single payer health care are also likely to believe that human activity has not caused global warming. Makes no sense without the work of some cognitive bias.


Note: Only a member of this blog may post a comment.

Who links to my website?