Abstract
Humans evolved in the context of intense intergroup competition, and groups comprised of loyal members more often succeeded than groups comprised of nonloyal members. Therefore, selective pressures have sculpted human minds to be tribal, and group loyalty and concomitant cognitive biases likely exist in all groups. Modern politics is one of the most salient forms of modern coalitional conflict and elicits substantial cognitive biases. The common evolutionary history of liberals and conservatives gives little reason to expect protribe biases to be higher on one side of the political spectrum than the other. This evolutionarily plausible null hypothesis has been supported by recent research. In a recent meta-analysis, liberals and conservatives showed similar levels of partisan bias, and several protribe cognitive tendencies often ascribed to conservatives (e.g., intolerance toward dissimilar other people) were found in similar degrees in liberals. We conclude that tribal bias is a natural and nearly ineradicable feature of human cognition and that no group—not even one’s own—is immune.
The human mind was forged by the crucible of coalitional conflict (Geary, 2005). For many thousands of years, human tribes have competed against each other. Coalitions that were more cooperative and cohesive not only survived but also appropriated land and resources from other coalitions and therefore reproduced more prolifically, thus passing their genes (and their loyalty traits) to later generations (Tooby & Cosmides, 2010). Because coalitional coordination and commitment were crucial to group success, tribes punished and ostracized defectors and rewarded loyal members with status and resources (as they continue to do today). Thus, displays of loyalty and commitment to other members of the tribe also enhanced individual-level fitness (by increasing status and resources and minimizing risks of ostracism). Over time, this practice would select for traits that signal and enhance coalitional commitment (Berreby, 2005), such as in-group favoritism (Tajfel & Turner, 1979). Tribalism, therefore, is natural. 1
Tribal Bias
Although tribal loyalties inspire many noble behaviors, they can impel humans to sacrifice sound reasoning and judgmental accuracy for group belonging and commitment (Kahan, Peters, Dawson, & Slovic, 2017). In other words, tribal loyalties can lead to tribal biases. For example, people selectively approach information that supports their tribe’s interests and avoid information that has potential to harm their tribe (by watching particular news networks or forming “echo chambers” in their social environments; Stroud, 2010). Also, people evaluate information they are exposed to in a biased manner by being uncritically accepting of information that supports their tribe’s agenda and more skeptical of information that opposes it (Ditto, Liu, et al., 2019). These kinds of cognitive biases are problematic for two reasons. First, post-Enlightenment societies prize reason and rationality and no longer explicitly tolerate obvious displays of in-group favoritism. And second, modern governments require the coordination of multiple groups (e.g., political groups) to function. Biases decrease the likelihood of consensus as groups fail to agree even on the facts in a particular debate.
There are at least two reasons that tribalism distorts beliefs. First, beliefs display and signal loyalty to group goals. Asserted opinions at least partially function as indicators of behavioral intentions and therefore as indicators of coalitional membership (Pietraszewski, Curry, Petersen, Cosmides, & Tooby, 2015). When one asserts “abortion is immoral,” one indicates willingness to coordinate with others to regulate abortion. Coalitions that generally oppose abortion (e.g., the modern Republican party) react negatively toward putative members who assert skepticism about prolife principles (Ditto & Mastronarde, 2009) because this attitude indicates an unwillingness to cooperate on that goal. If beliefs are held fervently, compel strong emotional displays, or are costly to hold, they might function as honest (and thus trustworthy) loyalty signals (Kurzban & Christner, 2011). Perhaps perversely, dogmatism and resilience to contrary evidence likely enhance the persuasiveness of the signal because they show that one is strongly dedicated to the group’s ideology in spite of potential consequences (e.g., being wrong about a difficult-to-answer question).
Second, beliefs are precursors to potential arguments that support the interests of the group, which coalitions are often formed to pursue and protect (e.g., wealthy people who want low tax rates). In modern societies, violence is verboten, so tribes prevail not by conquering other tribes but by persuading other people—often by making arguments. Sincere beliefs generally lead to better and more zealous arguments than cynical hypocrisy (von Hippel & Trivers, 2011). Therefore, people are motivated to favor and believe information that promotes their group’s interests and resist information that opposes their group’s interests because it makes them more persuasive proponents of their group’s cause (Kahan, Jenkins-Smith, & Braman, 2011).
Political Bias
These two reasons also likely explain why politics appears to be one of the most fertile grounds for bias (Van Bavel & Pereira, 2018). Political contests are highly consequential because they determine how society will allocate coveted resources such as wealth, power, and prestige. Winners gain control of cultural narratives and the mechanisms of government and can use them to benefit their coalition, often at the expense of losers. Given these high stakes, motivations to signal group loyalty and to defend the positions of the group are likely particularly powerful in politics.
Within the political domain, individuals appear most biased about issues that are most important to the group, which often include moral commitments (Ditto, Pizarro, & Tannenbaum, 2009). As noted above, moral commitments signal that one is willing to conform to the rules of the coalition. Therefore, groups are particularly prone to giving status to individuals who conform to and vocalize support for moral norms and deducting status from individuals who rebel and vocalize dissent against those norms (DeScioli & Kurzban, 2013). Thus, we can expect tribal biases to be especially large for important moral commitments (Tetlock, 2002). For example, if opposing abortion is a central goal for the political right, conservatives will be particularly biased about facts surrounding abortion. If enhancing the status of women in society is a central moral goal of the political left, liberals might be particularly biased about facts surrounding the gender wage gap.
However, humans also care about truth and accuracy (for obvious evolutionary reasons), and so biases are most likely to emerge for issues in which the truth is ambiguous (Munro, Weih, & Tsai, 2010). Many if not most political (and moral) disagreements are about ambiguous issues. Experts disagree about when a fetus or child can experience conscious pain and about the many contributors to the gender wage gap (and even the size of it). Even if experts could agree on the facts, political positions often reflect opinions about what ought to be the case (often subjective beliefs) based on beliefs about what is the case (ideally objective facts). For example, if the within-profession wage gap is largely due to women’s choices to work fewer hours, should they be paid the same as men? Policy choices often involve painful and complicated trade-offs (e.g., interfering with free-market autonomy to reduce income inequality, investing in new and more costly energy technology to minimize climate change).
When the truth is ambiguous, tribal biases are more powerful because argument is more important than when the truth is clear. Groups do not debate whether trees exist because the answer is virtually undeniable. They do, however, debate whether fetuses deserve various legal protections or whether women are paid less than men for equal work because there are intelligent arguments on both sides of these issues, and there is no one obvious correct answer. There is an unfortunate tribal logic here. One might imagine that ambiguity would compel humility and confessions of uncertainty, but when ambiguity occurs in the context of coalitional conflict, it may actually increase epistemic arrogance and bias. This is perfectly sensible, however, if we remember that humans are coalitional animals, not dispassionate reasoners. They were not “designed” to be humble; rather, they were “designed” to conform and to protect the status of their tribe (Kahan et al., 2017).
Our guiding assumption, then, is that tribal bias is a nearly ineradicable element of human nature and that it causes predictable cognitive biases (those that benefit the self and the group). Specifically, people will be biased in favor of their tribe, particularly for issues important to the tribe (often moral issues) and particularly when ambiguity is high and therefore the importance of argument and persuasion is high. Given that modern liberals and conservatives share evolutionary histories that favor loyalty signals and tribal biases, it is a priori likely that the psychological propensities for bias would be similar on the political left and right. We call this the evolutionarily plausible null hypothesis, and recent research has supported it.
Everyone Is a Little Bit Biased . . .
Social sciences for a long time focused especially on the biases of conservatives; some scholars argued that conservatives are more biased than liberals (e.g., Jost, Glaser, Kruglanski, & Sulloway, 2003). 2 But in recent years, researchers have pushed back against this narrative, contending that the overwhelming preponderance of liberals in the social sciences may have skewed research about political ideologies and the people who hold them. Liberals likely see their own biases as truths (Pronin, Lin, & Ross, 2002) and see conservative beliefs as peculiar and wrong; therefore, they seek to explain the “conservative mind” and its perplexing biases (Duarte et al., 2015; Eitan et al., 2018).
This insight inspired Ditto, Liu, and colleagues (2019) to conduct a meta-analysis to test these competing hypotheses. They examined 51 experiments that tested the tendency for liberals and conservatives to evaluate identical information more favorably when it supports their own political commitments than when it opposes them (e.g., a death-penalty supporter evaluating scientific methods as more valid when the results of those methods support rather than oppose the deterrent efficacy of the death penalty). They found strong support for the symmetry hypothesis: Liberals and conservatives were both biased and to virtually equal degrees. Because the included studies were performed under tightly controlled laboratory conditions, these results cannot tell us how liberal and conservative biases might vary over time and context, but they do suggest that liberals and conservatives share the same basic psychology that leads to bias—and to similar degrees. This finding is consistent with the evolutionarily plausible null hypothesis: Tribal bias is natural, and thus all political tribes should be similarly susceptible to it.
. . . Even Liberals
Whereas earlier scholars often emphasized that conservatives were higher in proclivities that ought to predict stronger biases (compared with liberals), such as authoritarianism and dissonance avoidance, a new wave of research in social psychology suggests that many of these proclivities exist in equal levels in conservatives and liberals. As can be seen in Table 1, these proclivities include authoritarianism, discrimination, dissonance avoidance, prejudice, selective exposure, and resistance to science. For example, although researchers previously thought conservatives were more intolerant of dissimilar other people, such results may have been due to confounds between the target groups investigated by liberal researchers (e.g., African Americans) and the political ideology of the target groups (e.g., African Americans tend to be politically liberal). More recent work suggests that people exhibit higher intolerance toward groups perceived as more dissimilar to their own group and that both liberals and conservatives exhibit this bias to a similar degree (Brandt, Reyna, Chambers, Crawford, & Wetherell, 2014).
Claims From Recent Work Demonstrating More Symmetry Between Liberals and Conservatives Than Previously Believed
This does not mean that conservatives and liberals are similar in all ways or that one group will never be vastly more biased or incorrect than the other—they will (Ditto, Clark, et al., 2019; Federico & Malka, 2018). Groups, as we have argued, are most biased about issues that are morally important and ambiguous. The general psychological propensities for bias appear similar on the political left and right, but there are predictable domain-specific asymmetries in bias.
Consider a few examples. Conservatives appear more motivated to reject anthropogenic climate change than liberals, likely because a belief in climate change seems to support government regulation and more centralization and hurts the fossil fuel industry, an important part of the Republican base in the United States (Lewandowsky & Oberauer, 2016). Conservatives may also exaggerate the amount of choice people exercise over their sexuality because homosexuality is considered immoral by a substantial proportion of the religious believers in the Republican coalition (Haider-Markel & Joslyn, 2008), and contending that it is a free decision rather than an innate inclination is more compelling for moral condemnation (Clark, Baumeister, & Ditto, 2017). On the other hand, a growing body of work suggests that liberals in general are more biased than conservatives about traditionally conceived disadvantaged groups (e.g., women, Blacks; see Table 2), likely because an important moral value of the political left is opposition to inequality (Jost, Nosek, & Gosling, 2008).
Findings From Recent Work Documenting That Liberals Show More Bias Than Conservatives in Domains Involving Disadvantaged Groups
Note that if one group currently has more or stronger concerns (because of historical and time-variant factors such as rapidly changing demographics or having recently lost a presidential election) or if one group has more moral convictions in general, one might predict more bias in that group (during that time period or in general). However, our current best estimate is that domain-specific asymmetries between liberals and conservatives appear to produce general symmetries in protribe biases among liberals and conservatives when averaged across multiple domains (and over at least a brief period of time). Until newer or better information contradicts these recent findings, it seems reasonable to posit that liberals and conservatives are roughly symmetrical in their protribe cognitive tendencies.
Conclusion
Humans are tribal creatures. They were not “designed” to reason dispassionately about the world; rather, they were “designed” to reason in ways that promote the interests of their coalition (and hence, themselves). It would therefore be surprising if a particular group of individuals did not display such tendencies, and recent work suggests, at least in the U.S. political sphere, that both liberals and conservatives are substantially biased—and to similar degrees. Historically, and perhaps even in modern society, these tribal biases are quite useful for group cohesion but perhaps also for other moral purposes (e.g., liberal bias in favor of disadvantaged groups might help increase equality). Also, it is worth noting that a bias toward viewing one’s own tribe in a favorable light is not necessarily irrational. If one’s goal is to be admired among one’s own tribe, fervidly supporting their agenda and promoting their goals, even if that means having or promoting erroneous beliefs, is often a reasonable strategy (Kahan et al., 2017). The incentives for holding an accurate opinion about global climate change, for example, may not be worth the social rejection and loss of status that could accompany challenging the views of one’s political in-group.
However, these biases decrease the likelihood of consensus across political divides. Thus, developing effective strategies for disincentivizing political tribalism and promoting the much less natural but more salutary tendencies toward civil political discourse and reasonable compromise are crucial priorities for future research. A useful theoretical starting point is that tribalism and concomitant biases are part of human nature and that no group, not even one’s own, is immune.
Recommended Reading
Ditto, P. H., Liu, B. S., Clark, C. J., Wojcik, S. P., Chen, E. E., Grady, R. H., . . . Zinger, J. F. (2019). (See References). A meta-analysis of partisan-bias studies (which found that liberals and conservatives showed an equivalent tendency to evaluate politically congenial information more favorably than politically uncongenial information), including a discussion of how to reconcile conflicting literature on the question of symmetry in partisan bias.
Eitan et al. (2018). (See References). An article demonstrating the extent to which research in political social psychology can be affected by liberal viewpoints and values.
Federico, C. M., & Malka, A. (2018). (See References). Example of a review article that challenges the notion that conservative ideology is invariably linked with certain psychological dispositions and argues instead that the association is often dependent on various factors, such as issue, context, and group loyalty.
Kahan, D. M., Peters, E., Dawson, E. C., & Slovic, P. (2017). (See References). A useful article for understanding how motivated reasoning in politics serves to maintain individuals’ standing in important in-groups (e.g., on the basis of political identity).
Van Bavel, J. J., & Pereira, A. (2018). (See References). A comprehensive and topical overview on ways in which partisan identity can affect individuals’ cognition, judgments, and decision making.
