The danger of myside bias

Fighting bad thinking is an uphill battle some days.

I'm very much including myself in this assessment.  I have biases and preconceived notions and places where I stumble just like everyone else.  Fixing these errors would be nice -- can you imagine the world if all of us were able to think clearly and make our decisions based on evidence?

A pipe dream, I know, and all the more so after I read a new paper in Journal of Cognitive Psychology called "My Point is Valid, Yours is Not: Myside Bias in Reasoning About Abortion," by Vladimíra Čavojová, Jakub Šrol, and Magdalena Adamus of the Institute of Experimental Psychology at the Slovak Academy of Sciences in Bratislava, Slovakia.

In an elegant piece of research, Čavojová et al. gave a series of logic puzzles to volunteers who had been asked in a prior questionnaire what their attitudes toward abortion were, and whether they had prior experience with formal logic.  They were then asked to determine whether various syllogisms were valid or not.  Some were neutral:

All mastiffs are dogs.
Some mastiffs are black.
Therefore, some of the things that are black are dogs. 

(Valid)

Some had to do with abortion:

All fetuses are human beings
Some human beings should be protected.
Therefore, some of those who should be protected are fetuses. 

(Invalid)

To solve each of the syllogisms, it should be irrelevant what your opinion on abortion is; the rules are that if the premises (the first two statements) are true, and the argument is valid, then the conclusion is true.  The participants were told from the outset to treat the premises as true regardless of their views.

Gregor Reisch, Logic Presents Its Main Themes (ca. 1505) [Image is in the Public Domain]

What is fascinating is that both people who were pro-choice and pro-life had a hard time rejecting invalid syllogisms that gave them a conclusion they agreed with, and accepting valid syllogisms that gave them a conclusion they disagreed with.  This pattern held equally with people who had training in formal logic and those who did not.  It's as if once we're considering a strongly-held opinion, our ability to use logic goes out the window.

The authors write:

The study explores whether people are more inclined to accept a conclusion that confirms their prior beliefs and reject one they personally object to even when both follow the same logic.  Most of the prior research in this area has relied on the informal reasoning paradigm; in this study, however, we applied a formal reasoning paradigm to distinguish between cognitive and motivational mechanisms leading to myside bias in reasoning on value-laden topics (in this case abortions).  Slovak and Polish (N = 387) participants indicated their attitudes toward abortion and then evaluated logical syllogisms with neutral, pro-choice, or pro-life content.  We analysed whether participants’ prior attitudes influenced their ability to solve these logically identical reasoning tasks and found that prior attitudes were the strongest predictor of myside bias in evaluating both valid and invalid syllogisms, even after controlling for logical validity (the ability to solve neutral syllogisms) and previous experience of logic.

Which reinforces my not very optimistic notion that however good our brains are, humans remain primarily emotional creatures.  When something elicits a strong emotional response, we're perfectly willing to abandon reasoning -- and sometimes aren't even aware we're doing it.

All of which bodes nothing good from any attempt to correct these errors.  As we've discussed before, even trying to combat bad thinking initiates the backfire effect, wherein people tend to double down on beliefs if they're challenged, and even if they're given concrete evidence that they're wrong.

Makes you wonder what I think I'm accomplishing by writing this blog, doesn't it?

It's not futile, however; however this emotional bent is impossible to eradicate, you can adjust for it as long as you know it's there.  So I suppose this research should give us hope that even if we can't think with perfect clarity all the time, we can at least move in the right direction.
 **************************************

This week's Skeptophilia book recommendation is from the brilliant essayist and polymath John McPhee, frequent contributor to the New Yorker.  I swear, he can make anything interesting; he did a book on citrus growers in Florida that's absolutely fascinating.  But even by his standards, his book The Control of Nature is fantastic.  He looks at times that humans have attempted to hold back the forces of nature -- the attempts to keep the Mississippi River from changing its path to what is now the Atchafalaya River, efforts in California to stop wildfires and mudslides, and a crazy -- and ultimately successful -- plan to save a harbor in Iceland from a volcanic eruption using ice-cold seawater to freeze the lava.

Anyone who has interest in the natural world should read this book -- but it's not just about the events themselves, it's about the people who participated in them.  McPhee is phenomenal at presenting the human side of his investigations, and their stories will stick with you a long time after you close the last page.

[If you purchase the book from Amazon using the image/link below, part of the proceeds goes to supporting Skeptophilia!]

Magnets, politics, and preconceived notions

Two stories showed up just in the last couple of days that are interesting primarily in juxtaposition.

First, we had a scholarly paper published in PLOS One, entitled "Copper Bracelets and Magnetic Wrist Straps for Rheumatoid Arthritis – Analgesic and Anti-Inflammatory Effects: A Randomised Double-Blind Placebo Controlled Crossover Trial."  In it, we find out what most skeptics suspected from the get-go -- that magnetic and copper bracelets and anklets and necklaces and shoe-sole inserts and so on are a complete non-starter when it comes to treating disease.

These claims have been around for years, and usually rely on pseudoscientific bosh of the kind you find in this site, wherein we have the following "explanation:"

Life developed under the influence of the earth's geomagnetic field.  We are surrounded by a sea of magnetism.  The human body, its individual organs and each of the millions of cells making up the organs and the body bathed by this sea are magnetically charged.  Cell regulation, tissue function and life itself are controlled by internal electromagnetic currents.  In disease states, these electromagnetic potentials are altered but fortunately can be favorably influenced by the external application of magnetics...  Used correctly, Electro-Magnetic Energy Fields are a proven therapeutic modality.  Research and clinical experience has established that the very gentle, EULF, low power pulsed magnetic energy improves the repair of damaged tissue and reduction of pain, improved oxygen transport in the red blood cells, increased nutrient and oxygen uptake at the cellular level.  Greater elasticity of blood vessels, changes in acid/alkaline balance, altering of enzyme and hormone activity, all play an important role in the return to good health...  Negative magnetic fields oxygenate and alkalize by aiding the body's defense against bacteria, fungi, and parasites, all of which thrive in an acid medium.  In degenerative diseases, calcium is found deposited around inflamed joints, bruised areas on the hell, and in bones and kidney stones.  Infections occur because they function well in an acidic, oxygen deficient state.

Which, in my opinion, should win some kind of award for packing the most bullshit into a single paragraph.

So the whole copper-and-magnet thing never did make much sense.  But don't take my word for it; here's what Richardson, Gunadasa, Bland, and MacPherson said, after having run a double-blind efficacy test on magnetic bracelets:

The results of this study may be understood in a number of ways. The most obvious interpretation is that they demonstrate that magnetic wrist straps, and also copper bracelets, have little if any specific therapeutic effects (i.e. beyond those of a placebo) on pain, inflammation, or disease activity in rheumatoid arthritis...  The fact that we were unable to demonstrate... a difference for the primary outcome measure on its own, nor indeed any of the other core measures employed, strongly suggests that wearing magnetic wrists straps, or copper bracelets, in order to minimise disease progression and alleviate symptoms of rheumatoid arthritis is a practice which lacks clinical efficacy.

But as I said, this is hardly a surprise to skeptics, who doubted the whole thing pretty much from the outset.

The second story at first seems to connect to the first in only a tangential fashion at best.  Chris Mooney, a skeptical writer of well-deserved high reputation, wrote about it this week in Grist in a piece called "Science Confirms: Politics Wrecks Your Ability to do Math."   In Mooney's article we hear about a study by Dan Kahan and his colleagues, of Yale Law School, in which two groups of people were asked to solve the same (rather difficult) mathematical problem -- but one group was given the problem in the context of its being about "the effectiveness of a new skin cream for rashes," and the other group that it was about "the effectiveness of a new law banning private citizens from carrying concealed handguns in public."

What Kahan's study found was that when the problem involved the relatively emotionally-neutral context of a skin cream, your ability to solve the problem correctly depended upon only one thing -- your skill at math.  In other words, both Democrats and Republicans scored well on the problem if they were good at math, and both scored poorly if they were bad at math.  But when the problem involved handguns, a different pattern emerged.  Here's how Mooney explains the results:

So how did people fare on the handgun version of the problem? They performed quite differently than on the skin cream version, and strong political patterns emerged in the results — especially among people who are good at mathematical reasoning. Most strikingly, highly numerate liberal Democrats did almost perfectly when the right answer was that the concealed weapons ban does indeed work to decrease crime...  an outcome that favors their pro-gun-control predilections. But they did much worse when the correct answer was that crime increases in cities that enact the ban... 

The opposite was true for highly numerate conservative Republicans: They did just great when the right answer was that the ban didn't work... but poorly when the right answer was that it did. 

Put simply: when our emotions and preconceived notions are involved, data and logic have very little impact on our brains.

This is a profoundly unsettling conclusion, especially for people like me.  Every day I get up and write about how people should be more logical and rational and data-driven, and here Kahan et al. show me that all of the double-blind studies in the world aren't going to convince people that their magnet-studded copper bracelets aren't helping their arthritis pain if they already thought that they worked.

It does leave me with a sort of bleak feeling.  I mean, why test wacko claims, if the only people who will believe the results are the ones who already agreed with the result of the experiment beforehand?  Maybe this justifies the fact that I spend as much time making fun of woo-woos as I do arguing logically against them.  Appeal to people's emotions, and you're much more likely to get a result.

On the other hand, this feels to me way too much like sinking to their level.  I live in hope that the people who are convinced by what I write -- and maybe there have been a few -- have been swayed more by my logic than by my sarcasm.

But given human nature -- and Kahan's experiment -- maybe that's a losing proposition.