Understanding Confirmatory Bias and Dead End Arguments

By Gary Berg-Cross

In her popular blog “General Myers and His Endless War on Error”, Sarah Hippolitus took on PZ Myers essay, Sunday Sacrilege: Sacking the City of God, for its deliberate and provocative vitriol. The argument is simple, telling people they are wrong in this way just doesn’t work and is perhaps counterproductive. Sarah cited Chris Mooney's essay, The Science of Why We Don't Believe Science, in which Mooney explains why berating people, such as religious conservatives who are emotionally committed to an idea, can make them defensively cling harder to their beliefs and biases. An in-your-face criticism raises emotions and people respond defensively overwhelming any chance at reason. Over the last 50 years of social and cognitive research we have learned a considerable amount of the general type of phenomena of how people frame and defend biases (bracketology being one area where “experts” are visibly wrong) and unsupported beliefs and how we rationalization our beliefs rather than use reason to arrive at opinions. I’ve written on this topic earlier in such blogs as “Rationalizing Irrational Choices: That $45 entree and presidential choices” and Boxing Ourselves In with Category Errors. We have come to a sad, but scientific understanding from a coordinated system of studies that we humans are imperfect reasoners. The fact is that the emphasis on rationality encouraged by Enlightenment thinkers is a noble effort, but difficult, and does not come easily in many circumstances. The modern parlance covering this is to talk about gut feeling that overwhelm and, in effect, reverse engineer our opinions. More recently the studies have focused on group differences in cognitive styles and the role of emotions, like fear, in these differences.

The scientific study of general phenomena of changing the minds of people with convictions, especially faith-based ones,  goes back a long way, but progress was made starting in the 50s. Stanford University social psychologist Leon Festinger summarized it succinctly based on a study of a UFO cult that was convinced the world would end on December 20^th 1954 (see“When Prophecy Fails”):

"A man with conviction is a hard man to change. Tell him you disagree and he turns away. Show him facts or figures and he questions your sources. Appeal to logic and he fails to see your point."

And it wasn’t just odd folks that show biases responses. Experimental studies of experts, such as Physicians, show they are often biased in clinical judgments. Their expertise could be improved upon by simple statistical rules, a discovery that prompted some hope for things like artificial, expert systems to aid human judgment.But this human tendency gets enhanced when it interacts with factors like ideology,  frightening circumstances and cultural framing.

The more recent studies on differences of reasoning style, which I touched on in Epistemological Styles. There I contrasted scientific with theist styles. Scientific styles are cognitively taxing and require considerable cultural support for chained combinations of postulates that get tested by experiment. These may confirm a belief but also disconfirm it. It’s all part of a humbling style that values exploration of ideas with controlled observation and measurement, analogical models and statistical analysis of regularities of populations.Such methods have been applied to self-study processes by which small groups have moved the country  away from intellectual  processes to more of a conservative agenda in this  country (see for example Hacker and Pierson's work Off Center). Such conservative, social-political engineering reflect reaching people who prefer far simpler style . It is to be an idea advocate and largely focus on evidence to justify existing belief and gut feelings.Taken as a whole this tends to appeal to conservative minds more than liberal ones.

Chris Mooney, author of The Republican War on Science, discusses such things in his recent book The Republican Brain The Science of Why They Deny Science--and Reality. His finding is simply states: conservatives and liberals don’t just have different ideologies; they have different psychologies (aka epistemological styles) that help explain their fact and expert denying positions on such things as climate change and evolution. It’s not just that they don’t know facts it is that they have a tendency towards a confirmatory bias that reflects a preference to focus on evidence that supports existing belief. Another part of who conservatives are is their preference for ideas that bind groups together rather than “truth” in an objective sense. And one of the big factors driving conservative beliefs is their tendency to emotional, fight or flight responses. Eye tracking studies, for example, show that people who score high on conservative positions tend to track “threat” objects in an environment (a knife or gun – maybe a stranger). There’s survival value in such a tendency so one may imagine how such a tendency was selected in populations and later frozen into cultural values. Conservatives tend to be ready to be afraid and this can turn off the reasoning side of the brain.
You can see a good discussion of all of this on the recent Up With Chris Hayes, where Mooney and Jonathan Haidt, UVA moral psychologist & author of The Righteous Mind, discussed group differences. They agreed that liberals tend to be more open to new experiences, new data and convincing finds. As a consequence they tend to be more sympathetic to scientific process, and take their scientific findings more seriously. This is a tendency/predisposition and not a hard and fast thing in all circumstances.  The socially generated tendency of Conservatives, meanwhile, is that they just do it differently. Haidt explained:

“I want to fully agree with Chris that the psychology does predispose liberals more to be receptive to science; my own research has found that conservatives are better at group-binding, at loyalty, and so if you put them in a group-versus-group conflict, yes, the right is more prone, psychologically, to band around and sort of, circle the wagons.”

This is consistent with Sarah’s earlier argument about the reaction of believers to strong, mocking arguments. In conversations with conservative thinkers (as opposed to  liberal styles) we should be aware that they may not have the same regard for the Enlightenment’s style of rational argument. We are either fooling ourselves or just being a bit too rigid in our style. But we need not be quiet or withdrawn certainly in our own communuty, and we can use the new understandings that come from cognitive and social science to guide us. The value of framing (see my Framing Arguments: You say Flaming Atheists and I Say Non-Confrontational Humanist) 
is one tool we can use. Another is to recognize that individual reasoning is biased. We are often battling a strong confirmatory bias that repels facts and strict logic. What we need instead is get into a dialog to challenge our biases as well as others. Socrates was on to this a long time ago and its time to update that technique drawing lessons from the relevant science. Overcoming a reflexive conservative denial of the science of denial would be a big step. We need to get the word out on that.