Monday, January 15, 2018

Consider counter-evidence to avoid bias

By Mathew Goldstein

Neurologists at the Brain and Creativity Institute at the University of South Carolina (USC) Dornsife College of Letters, Arts and Science watched the brains of 40 self-declared liberal students in a functional MRI. USC neuroscientists compared whether, and how much, people change their minds on non-political and political issues when provided counter-evidence. During their brain imaging sessions, participants were presented with eight political statements that they had said they believe just as strongly as a set of eight non-political statements. They were then shown five counter claims that challenged each statement.

Participants rated the strength of their belief in the original statement on a scale of 1-7 after reading each counter claim. The scientists then studied their brain scans to determine which areas became most engaged during these challenges.  Participants did not change their beliefs much, if at all, when provided with evidence that countered political statements. But the strength of their beliefs weakened by one or two points when provided with evidence that countered non-political statements.

The study, which concluded last month, found that people who were most resistant to changing their beliefs had more activity in the amygdala (a pair of almond-shaped areas near the center of the brain) and the insular cortex, compared with people who were more willing to change their minds. “The amygdala in particular is known to be especially involved in perceiving threat and anxiety,” said USC Psychologist Kaplan, explaining that “the insular cortex processes feelings from the body, and it is important for detecting the emotional salience of stimuli. That is consistent with the idea that when we feel threatened, anxious or emotional, then we are less likely to change our minds.” He also noted that a system in the brain called the default mode network surged in activity when participants’ political beliefs were challenged. “These areas of the brain have been linked to thinking about who we are, and with the kind of rumination or deep thinking that takes us away from the here and now,” Kaplan said.

People will flexibly react to changes in their environment. If a sidewalk or road is blocked then we have no difficulty understanding that we need to consider finding a different route to our destination. But we are not consistently rationally flexible, particularly with regard to beliefs that we link to our self-identity. Instead of prioritizing best fit with the overall available evidence, we may negatively react to evidence that conflicts with our self-identity linked beliefs similar to the way we negatively react to a threat.

People tend to link their religious beliefs to their self-identity at least as much as they do their political beliefs and they also may link their religious and political beliefs together. This is one reason why we should be careful about how we go about justifying our beliefs. We need to be careful to open-mindedly allow the overall available empirical evidence dictate to us what our beliefs about how the universe functions should be. We are prone to reversing this sequence and telling the universe how it functions as if we are each master of the universe deities. The universe is not about us, so what we think should be true, or what we want to be true, or how we define our self-identity, are irrelevant.

To try to avoid this error, my advice to everyone, regardless of whether you are a metaphysical naturalist or supernaturalist, is to consider what would need to be different about our universe to convince you to change your conclusion. Too often, when I ask this question I get pushback directed against the question itself. Not all atheists are empiricists. People react negatively to the question, claiming that it is wrong to talk about an alternative universe, that it is wrong to consider other possibilities, because that is a place of falsehood. It is said that our universe is naturalistic because supernaturalism is impossible, and to even ask such a question is to accept that supernaturalism is possible and thus is a mistake. It is said that supernaturalism is a non-starter and to even entertain it as a possibility is an unwarranted concession.

My response is this: We cannot trust our intuition, or anything mostly rooted in intuition, like faith or hope, to answer the big questions about how the universe functions because the answers to the big questions are mostly non-intuitive and counter-intuitive. So it is a mistake to rule out anything a-priori or to rely only on logic not anchored in evidenced. It is often inconsistent for some assertion to be simultaneously true and false. Therefore it is reasonable to conclude that given that X (naturalism) is true it is probably also the case that the opposite of X (supernaturalism) is impossible. But the impossibility of X being false when it is true is not a proper justification for concluding X is true, we still must justify our conclusion regarding X. To justify the conclusion that it is impossible for X to be false, we paradoxically should consider what is missing that would be required to properly justify a conclusion that X is false.

A-priori ruling out even identifying what qualifies as missing evidence favoring alternative conclusions is bad epistemology. Fairly considering what is needed to justify a conclusion entails also considering what would be needed to justify a contrary conclusion. Our justification for reaching a particular conclusion about how the universe functions is incomplete if we cannot identify missing justifications for concluding otherwise. When a conclusion is consistently supported by an abundance of highly diversified, interconnected, and direct empirical evidence it becomes unlikely that the available evidence will change so drastically as to favor the contrary conclusion, so we need not worry that our beliefs will be unstable if we allow the evidence to dictate. When a conclusion is inconsistently supported by rare, narrow, unconnected, and indirect non-empirical evidence then we should not have a strong commitment to that conclusion. Either way, there is no harm in identifying what evidence is missing that would change our conclusion if it was found.

Academic endeavors like science are publicly funded and some Senators and Representatives are prone to threaten to cut funding if they think scientific outputs interfere with their preferred political ideology. Elected school boards make decisions regarding educational curriculums, and elected governments decide if they fund private schools that set their own curriculums. Theism is a popular and often strongly held belief and educators and scientists fear popular antagonism if science is perceived as being anti-theistic. Theists falsely claim that science has a built in bias favoring naturalism, that science has a built-in self-dependency upon naturalism, and therefore science cannot fairly adjudicate the naturalism versus supernaturalism question. Some educators and scientists, many of whom are themselves theists, actively promote this false claim of bias at least in part because it is convenient as a means for avoiding provoking theists. Yes, modern knowledge favors naturalism. But our process of acquiring knowledge is not the source of this bias, the source of this bias is the nature of our universe.

People who imagine themselves living in a supernatural universe are not going to then respect a belittled empiricism that is deemed to lack the ability to challenge theism (note that most of the same theists would probably enthusiastically cite a scientific consensus that prayer works as a confirmation of God). With the false claim that empiricism has a built in bias for naturalism widely accepted it can be small additional steps to conclude that empiricism is similarly biased in multiple other contexts, that empiricism is not the best way to determine how the universe works, and that religion, wealthy business, popular entertainment, and political leaders are the most reliable sources of information about how the universe functions. Not all theists generalize away empiricism, expertise, and modern knowledge this way. But it appears that enough people generalize like this to cause mischief. Today we have wealthy businessman President Trump, maybe in 2020 it will be wealthy talk show host President Winfrey?

No comments: