Oct 10 2013

Extreme Dogmatism

It is the standard skeptical narrative that people are biased in numerous ways. The “default mode” of human behavior is to drift along with the currents of our cognitive biases, unless we have critical thinking skills as a rudder or paddle (choose your nautical metaphor). Metacognition – thinking about thinking – is the only way for our higher cognitive function (evidence, analysis, logic) to take control of our beliefs from our baser instincts.

Political ideology is one form of such bias. Psychologists have demonstrated that people generally will identify with a stated belief, and then will defend their existing belief by default simply because it’s theirs. This phenomenon seems to be exacerbated by ideology – identifying with a suite of beliefs that come as a package deal, with a convenient label.

One type of ideology is political. In the US this is usually thought of as a dichotomy between conservative and liberal, represented by the Republican and Democratic parties respectively. In reality the political landscape is more complex. Libertarians, for example, are economically conservative but socially liberal, because in reality they care about something else entirely, something tangential to the typical conservative-liberal axis, and that is personal freedom over government control.

Within the conservative-liberal paradigm, however, there has been an ongoing debate about which side, if either, is more dogmatic, anti-science, irrational, or biased. Chris Mooney in The Republican Brain argues that conservatives are more authoritarian by nature, and that explains why they tend to be more dogmatic even to the point of being anti-science.

The alternate hypothesis, however, is that it is really extremism that drives dogmatism (defined as holding your opinions to be objectively correct or superior). In this view extreme liberals can be just as dogmatic as extreme conservatives.

A recent psychological study looks directly at that controversy – Feeling Superior Is a Bipartisan Issue Extremity (Not Direction) of Political Views Predicts Perceived Belief Superiority, by Toner et al, published in Psychological Science.

In the study the researchers distinguish between dogmatism (the arrogant assertion of one’s opinions as facts) and belief superiority (holding that one’s specific opinions are superior to others). They questioned 527 Americans on issues that tend to be politically contentious, like immigration and abortion, and found that dogmatism (evaluated on a standard scale) was indeed higher among conservatives. Belief superiority (assessed for each individual belief), however, correlated with extremism, and not conservative vs liberal. They did also find a correlation between belief superiority and dogmatism, which makes sense.

There does seem to be a pattern in recent studies that conservatives tend to be more dogmatic and authoritarian. On this point I think that Mooney and others are onto something. At the same time there appears to be basic human psychological forces at work that transcend political ideology. This study supports that, noting that it is extremism, and not political orientation, that really correlates with hardened and irrational views.

As with all psychological studies, the real complexity is in the interpretation. There are two basic levels on which the interpretation of this kind of data is tricky. The first is that we may be seeing the influence of topic choice and current culture rather than something inherently different about the conservative vs liberal mind.

In other words, if you choose to study topics about which conservatives are more passionate, this may introduce an artifact that makes conservatives look more dogmatic and extreme. If you question conservatives about abortion, and liberals about nuclear power, you may find that conservatives are more dogmatic – but only because they care more about abortion than liberals do about nuclear power. In other words, the difference in intensity may be at the topic level, and not inherent to the political spectrum.

This is my main problem with Mooney’s work. I would like to see further replication across a broader range of issues, and some measure of topic intensity to see how that variable is affecting the outcome.

The other variable that comes into play is recent American culture. I call this the Fox News effect. My subjective experience is that conservative outlets like Fox News have had a radicalizing effect on Republicans, pushing them in an extreme direction.

It also seems to be true that the party that is out of office tends to be more passionate. Some liberals during the Bush administration were radicalized by their opposition to Bush. There was even a term invented for this – post election selection trauma (remember that one?).

In order for data on inherent differences between liberal and conservative to be compelling, therefore, we would need to see data across a wide spectrum of issues, some measure of issue intensity, and data across decades to measure the influence of current politics (such as which party is in the White House).

So – while I am prepared to believe that there are differences between conservatives and liberals that go deeper than their specific political opinions, including conservatives being more authoritarian and dogmatic, we need to be cautious before making any sweeping conclusions. We need to tease apart a host of influences and artifacts.

Meanwhile, the psychological literature is fairly consistent in demonstrating that people in general labor under a host of cognitive biases. Further, extremism (rigidly holding strong opinions) seems to be the main problem, rather than the opinions themselves. It is ideology itself, and not the details of a specific ideology, that renders people irrational.

In many ways, metacognition, critical thinking, and skepticism are about divesting oneself of ideology in favor of logic and evidence.

59 responses so far