My theory of basic human values strongly supports the fundamental claim of this article. It also helps to address some questions that Hibbing et al. raise.

Substantial cross-national evidence demonstrates that specific basic values underlie conservative versus liberal political preferences (Schwartz et al., in press). The values likely to reflect the strength of a person's negativity bias are the ones most relevant to the political domain. Figure 1 below presents a circular motivational continuum on which ten basic values are organized. The three outer circles specify principles that underlie and account for the organization of the values in the center. Any two values may express compatible or opposing motivations. The closer two values are in the circle (e.g., tradition and conformity), the more compatible their motivations; the more distant (e.g., tradition and hedonism), the more their motivations conflict.

Figure 1. Circular motivational continuum of basic values.

Consistently, universalism and self-direction values relate to liberal preferences whereas security, conformity and tradition values relate to conservative preferences. I briefly define the core motivational goals of these values.

• Universalism: understanding, appreciation, tolerance, and protection for the welfare of all people and for nature.

• Self-Direction: independent thought and action – choosing, creating, exploring.

• Security: safety, harmony, and stability of society, of relationships, and of self.

• Conformity: restraint of actions, inclinations, and impulses likely to upset or harm others and violate social expectations or norms.

• Tradition: respect, commitment and acceptance of the customs and ideas that traditional culture or religion provide.

The outer circle in the figure specifies the organizing principle that underlies value relations and is most relevant to political preferences. Values in the bottom part of the circle are based in the need to avoid or control anxiety and threat and to protect the self. Conservation values (bottom left) emphasize avoiding conflict, unpredictability, and change by submission and passive acceptance of the status quo. Self-enhancement values (bottom right) emphasize overcoming possible sources of anxiety by controlling, dominating, or gaining admiration. In contrast, values in the top part of the circle are relatively anxiety free; they express growth and self-expansion. Self-transcendence values (top left) emphasize promoting the welfare of others. Openness to change values (top right) emphasize autonomous, self-expressive experience.

Conservation values, which consistently correlate with conservatism, are aimed at coping with negative features of the environment, with protecting the self. Power values, which also frequently correlate with conservatism, are aimed at coping with negative features too, but through dominance and control over people and resources. It is highly plausible that greater physiological and psychological sensitivity to negative stimuli would lead people to endorse these values. It is equally plausible that those who endorse the values associated with liberalism, universalism and self-direction, are free to do so because they are less sensitive to negative stimuli.

Relations of basic values to worries further support the idea that a negativity bias undergirds conservatism. Hibbing et al. note a seeming incongruity in conservatives' greater concern than liberals' with protection from criminals, pathogens, and out-groups but their lesser concern with poverty, accidental shootings, and environmental degradation. The distinction between what Boehnke et al. (Reference Boehnke, Schwartz, Stromberg and Sagiv1998) called micro and macro worries can resolve this incongruity. Micro worries concern threats to the self or in-group – personal or in-group health, safety, economic and social success. Macro worries concern threats to entities external to the self – problems in the society or world of poverty, disease and harm to the environment. Micro worries, which correlate most positively with power values and most negatively with universalism values, cause conservatives more distress. Macro worries, which correlate most positively with universalism values and most negatively with power values, cause liberals more distress. Thus, what troubles people more depends on values that are grounded in self-protection versus growth, values that relate consistently to conservatism versus liberalism.

Are relations between negativity bias and ideology linear, or might the observed associations be a result of people who are located at the extremes on either or both of these dimensions? An individual's preference for security versus universalism values can serve as a proxy for the degree of the negativity bias. Using these values, I addressed the question of linearity with data from five rounds of the European Social Survey. Respondents from representative samples in 33 countries reported their values and located themselves on a 0–10 left–right scale.

In 18 of 20 Western democracies, an ANOVA revealed a linear pattern. The importance of security minus universalism was greater to the extent that respondents located themselves on the right rather than the left. In 14 of these countries, however, the extreme left (0) attributed slightly more importance to security than the less-extreme left (1). This supports the view that those at the extremes do not account for the relations of ideology to negativity bias. In contrast, in 13 post-communist countries, the pattern of greater importance for security minus universalism values on the right emerged in only four countries. Most likely, this reflects context specific meanings of ideologies and/or values (Piurko et al. Reference Piurko, Schwartz and Davidov2011; Thorisdottir et al. Reference Thorisdottir, Jost, Liviatan and Shrout2007); but one might wonder whether the experience of substantial threat and little autonomy might neutralize differences in sensitivity to negative stimuli.

Hibbing et al. avoid speculating about the amount of variance that negatively bias can explain in conservative versus liberal political stances. They locate negativity bias at an intermediate level of analysis and, as I understand them, argue successfully that it accounts for the coherence among an immense array of variables at higher levels of analysis. Yet, the question of how much influence negativity bias has on variables at higher levels is worth pursuing. This requires developing persuasive measures of this bias and studying its direct and mediated effects on personality variables known to predict political attitudes, ideology and voting. Basic values have emerged as perhaps the strongest personality variable to predict political thought and action (Caprara et al. Reference Caprara, Schwartz, Capanna, Vecchione and Barbaranelli2006; Schwartz et al., in press). Hence, assessing the link from negativity bias to values, assumed in this comment, would be a productive next step.