Hibbing et al. provide a masterful review of the literatures documenting psychological, behavioral, and physiological differences between liberals and conservatives. Moreover, they propose an elegant thesis to account for these literatures' main findings: that conservatives are more attuned and responsive to aversive stimuli than are liberals, and individual differences in dispositional negativity bias account for differences in conservative versus liberal political attitudes. We comment here on three key aspects of this thesis: the structure of political attitudes, the relations of negativity bias (as broadly conceptualized by Hibbing et al.) with social and economic political attitudes, and contextual moderators of the links between negativity bias and political attitudes. These considerations, we argue, suggest that dispositional influences on political attitudes, and therefore political conflicts, are narrower than the authors describe.

The authors define political conservatism versus liberalism as a single broad dimension, reflecting an “ancient and universal” political division pitting preferences for “stability and order” against those for “progress and innovation” (sect. 1, para. 1). The former side of this division is said to correspond with a “great many of the typical tenets of political conservatism,” including not only sociocultural preferences (such as support for restrictive immigration policies, harsh treatment of criminals, and social and moral traditionalism) but also economic preferences (such as opposition to government economic intervention and welfare programs) (sect. 6, para. 7).

However, individuals do not consistently organize their political attitudes along a broad conservative–liberal dimension (e.g., Treier & Hillygus Reference Treier and Hillygus2009). Rather, individuals with a high degree of political engagement – those who are most interested in and informed about political discourse – typically structure their political attitudes along a single dimension, but individuals who are less engaged typically do not (e.g., Zaller Reference Zaller1992). These findings present a challenge to any claim that a psychological disposition directly or organically leads individuals to adopt broad coherent packages of liberal or conservative attitudes. Such claims must contend with the fact that political attitudes do not organically cohere into particular packages; instead, the emergence of an overarching liberal-conservative dimension appears contingent upon exposure to political discourse, which itself emerges from the strategic actions of political elites.

The messy structure of political attitudes raises a second issue: whether negativity bias relates differently with different kinds of political attitudes. Hibbing et al. mention the claim that negativity bias influences economic attitudes less strongly than it does social attitudes (sect. 6.2, para. 3). They are right to do so, as a great deal of evidence is consistent with this view. For example, Smith et al. (Reference Smith, Oxley, Hibbing, Alford and Hibbing2011) found that physiologically assessed disgust sensitivity predicted conservative positions on certain sexuality-related issues but not on a single economic issue: welfare spending, tax cuts, free trade, or small government (p. 4, see also Inbar et al. Reference Inbar, Pizarro and Bloom2009a, pp. 720–21 and footnote 2). Oxley et al. (Reference Oxley, Smith, Alford, Hibbing, Miller, Scalora, Hatemi and Hibbing2008) found that physiologically assessed threat sensitivity predicted conservative position on a composite of sociocultural attitudes (e.g., the death penalty, gay marriage, abortion rights), but not a composite of economic attitudes, concluding that “physiological responses to threat are connected to socially protective policy positions but not to economic policy positions” (Supporting Online Material, p. 7). Self-report measures pertaining to negativity bias reliably predict conservative social attitudes but do not reliably predict conservative economic attitudes (Carney et al. Reference Carney, Jost, Gosling and Potter2008; Crowson Reference Crowson2009; Duckitt & Sibley Reference Duckitt and Sibley2009; Feldman & Johnston Reference Feldman and Johnston2014; Johnston Reference Johnston2013; Johnson & Tamney Reference Johnson and Tamney2001; Stenner Reference Stenner2005; Van Hiel & Mervielde Reference Van Hiel and Mervielde2004). To be sure, there are well-conducted studies showing some common dispositional correlates of economic and social attitudes (e.g., Gerber et al. Reference Gerber, Huber, Doherty, Dowling and Ha2010). However, the overall pattern of findings indicates that dispositional negativity bias reliably predicts social conservatism but does not reliably predict economic conservatism. Both social and economic positions are central aspects of right vs. left ideology, as well as frequent sources of political conflict. The available evidence indicates that conflicts about sociocultural issues – but not those about economic issues – may be organically rooted in individual differences in negativity bias.

This brings us to a third and final issue: the moderating effects of contextual variables. Hibbing et al. acknowledge that cultural and historical factors can influence links between negativity bias and political attitudes (sect. 3, para. 9). Another key contextual variable is exposure to political discourse. For example, one study found that high need for cognitive closure predicted more liberal (i.e., pro-government intervention) economic attitudes in both community and student samples, but predicted more conservative (i.e., anti-government intervention) economic attitudes in a sample of political activists; meanwhile, high need for closure predicted more-conservative sociocultural attitudes in all three samples (Golec Reference Golec2002). What might explain this pattern of results? We propose that individuals high in negativity bias are organically drawn toward conservative sociocultural attitudes. If such individuals become politically engaged (as in the case of activists), then they are frequently exposed to political discourse indicating that conservative sociocultural attitudes should be packaged with conservative economic attitudes, which they ultimately come to adopt. This study and others like it (e.g., Malka & Soto Reference Malka and Soto2011) suggest that some dispositional effects on political attitudes are not organic, but rather are contingent upon exposure to discursive messages about the packaging of political attitudes.

Taken together, the findings reviewed in this commentary support several conclusions about the relations between dispositional negativity bias and political attitudes. As proposed by Hibbing et al., individuals high in negativity bias do appear to be organically drawn toward conservative social attitudes such as restrictive immigration policies, harsh treatment of criminals, and social and moral traditionalism. However, such individuals do not appear to be organically drawn toward conservative economic attitudes. Instead, the effects of negativity bias on economic attitudes are contingent upon contextual factors such as culture and exposure to political discourse. More generally, the bottom-up structuring of sociocultural and economic attitudes compelled by dispositional influences might differ from, and even compete with, the top-down attitude structuring promoted by political discourse (e.g., Hatemi et al. Reference Hatemi, Eaves and McDermott2012). Conservative economic and sociocultural preferences do not seem to be bound together through the bottom-up influence of negativity bias.