By guest blogger Jesse Singal
If you follow mainstream science coverage, you have likely heard by now that many scientists believe that the differences between liberals and conservatives aren’t just ideological, but biological or neurological. That is, these differences are driven by deeply-seated features of our bodies and minds which exist prior to any sort of conscious evaluation of a given issue.
Lately, though, follow-up research has been poking some holes in this general theory. In November, for example, Emma Young wrote about findings which undermined past suggestions that conservatives are more readily disgusted than liberals. More broadly, as I wrote in 2018, there’s a burgeoning movement in social and political psychology to re-evaluate some of the strongest claims about liberal-conservative personality differences, with at least some evidence to suggest that the nature and magnitude of these differences has been overblown by shoddy or biased research.
Now, a new study set to appear in the Journal of Politics and available in preprint here suggests that another key claim about liberal-conservative differences may be less sturdy than it appears.
Many researchers have argued that people are more likely to endorse conservatism if they want to reduce feelings of threat in their lives, note the authors, led by Mathias Osmundsen at Aarhus University. “In this view, so-called ‘threat-sensitive’ individuals find the order inherent in a conservative ideology attractive,” they write.
Political scientists’ favored method of measuring threat-sensitivity in recent years has been so-called psychophysiological studies, which gauge how the body responds to what is going on in the mind. The key studies here measure participants’ skin conductance (also known as electrodermal activity, or EDA), using electrodes attached to the fingertips. This provides an index of automatic bodily arousal. A big turning point came in 2008, when Douglas Oxley and his colleagues found that conservatives had higher EDA than liberals when viewing threatening images, including pictures of spiders, maggots and guns, and several studies with similar findings have since been published.
For this new study, Osmundsen and colleagues sought to investigate the reliability of these results. The paper is a three-part effort consisting of a replication attempt, with a much larger sample size, on a group of Danes and Americans; a re-analysis of all the previously published studies on this subject with an eye toward ascertaining their quality; and a more zoomed-out evaluation of the instruments being used by researchers to do this sort of work.
What the team found suggests a field very much in need of some refinement. Some of the issues here are rather fundamental: For example, they write, “EDA is influenced by a range of factors that are likely to vary randomly and arbitrarily across individuals … includ[ing] outside noises, deep breaths, coughs, room temperature, bodily movements, thickness of the skin of the fingertips, pre-experiment arousal (e.g., from having biked to the lab) and so forth.” Researchers are simply assuming that measures of EDA are (more or less) good proxies for fear responses, ignoring a host of other factors that potentially introduce a great deal of noise. Because the extant literature so rarely indicates that these factors have been taken into account, the team writes, “we simply do not know if measures of EDA in political science research are empirically reliable.”
The literature fails to answer other important questions as well, argue the authors, like whether EDA responses to positive imagery mirror those to scary imagery, which would suggest it isn’t really fear being measured, at root, but arousal more generally. Indeed, a more recent theory has argued that conservatives may be more easily aroused than liberals, rather than more attuned to threat specifically.
The replication study was also an attempt to shed light on some of these questions by evaluating the reliability of the EDA measurement procedure itself, simply by comparing and correlating the participants’ EDA responses to various positive and negative stimuli.
The researchers found that their EDA measures were “very unreliable.” For example, there was almost no internal consistency when it came to responses to threatening (or to disgusting) stimuli, suggesting that whatever EDA is measuring, it isn’t a threat response. But the data were consistent with the idea that the EDA tests were measuring “individual differences in baseline physiological reactivity.” And although the researchers were able to replicate the correlation between conservatism and heightened EDA responses to negative stimuli in their American sample, there was no such correlation among Danish participants. (Another study out this week in Nature Human Behaviour failed to replicate the correlation among Dutch or American participants). This suggests that the original finding may not be generalizable to the wider population.
Interestingly, when the researchers simply asked the participants to gauge their explicit response to the imagery, they found sturdier correlations. “In contrast to physiological responses to threat, self-reported reactions to the threatening images were associated in both Denmark and the United States with more conservative beliefs, but only significantly so for the two measures that arguably reflect social [rather than fiscal] conservatism[.]”
So where does all this leave us? It appears that the tools used to measure fear-sensitivity in many previous studies might not be measuring that at all. This, of course, does not mean that there are no arousal-based differences between liberals and conservatives; it could be there are some good-sized and robust ones, for all we know. But the mixed results of the replication attempt suggest that even this weaker claim is far from certain. They also suggest that, while measures of automatic (or implicit) responses are very ‘in’ right now, they aren’t necessarily a better or more meaningful tool than simply asking people how they feel.
– The Psychophysiology of Political Ideology: Replications, Reanalysis and Recommendations [this study is a preprint meaning that it has yet to be subjected to peer review and the final published version may differ from the version on which this report was based]
Post written by Jesse Singal (@JesseSingal) for the BPS Research Digest. Jesse is a contributing writer at New York Magazine, and he publishes his own newsletter featuring behavioral-science-talk. He is also working on a book about why shoddy behavioral-science claims sometimes go viral for Farrar, Straus and Giroux.
At Research Digest we’re proud to showcase the expertise and writing talent of our community. Click here for more about our guest posts.