Figure 5. A model of moderated mediation shows that reappraisal attenuates the otherwise robust relationship between disgust sensitivity and purity concerns, thus leading to less support for conservative policies, Experiment 4.

All values are regression coefficients. Purity variable consists of the items that suggest purity as a moral concern. Solid lines indicate significant paths and dashed lines indicate non-significant paths. *p<0.05, **p<0.01, ***p<0.001. c indicates the conditional indirect effect of disgust sensitivity on support for conservative policies, whereas c′ indicates direct effect. (R) and (NR) denote reappraisal and non-reappraisal conditions respectively. Binary indicator variable for suppression was entered as a covariate. All beta coefficients (a through e) are unstandardized. Standard errors, p-values, and 95% confidence intervals for the corresponding coefficient estimates are as follows. a = 0.75, SE = 0.64, p<0.001. b = 0.72, SE = 0.10, p<0.001. c = −0.23 (R), SE = 0.21, p = 0.29, (−0.65, 0.29). c = 0.55 (NR), SE = 0.17, p = 0.002, (0.25, 0.91). c′ = 0.01, SE = 0.14, p = 0.97, (−0.29, 0.28). d = −1.07, SE = 0.37, p = 0.005. e = −0.01, SE = 0.19, p = 0.96.