Abstract:

Deep in our consciousness, we all know that we will die someday. The management of mortality and death-related emotions influences our behavior profoundly. The Terror Management Theory (TMT) and Death Reflection (DR) theory postulate different ways to manage Mortality Salience (MS), i.e., reminding people of their mortality. While the TMT proposes a series of external defenses to restrain the fear of mortality, the DR theory focuses on inner growth when contemplating our mortality. On the other hand, the role of emotion in managing mortality is unclear. Firstly, does MS lead to any change in emotion? Secondly, if MS does influence emotion, what are the main emotional reactions? Thirdly, what role do emotions play in MS effects on psychological constructures? We proposed the “Sorrow Management Theory” (SMT) based on the reflective function of sadness. We hypothesized that MS should induce both fear and sadness because death implies not only the destruction of the self but also the separation from the world. Correspondingly, fear and sadness should lead to two different modes of response to MS, i.e., “Terror Management” and “Sorrow Management”, respectively. Also, we hypothesize that fear should mediate the relationship between MS and external defenses, while sadness should mediate between MS and internal growth.

We conducted three experiments to compare “Terror Management” and “Sorrow Management”. In Experiment 1, we explored the emotional changes during and after MS, using various emotion measurement methods (i.e., physiological skin conductance response, heart rate, facial expressions, and self-report emotion). Eighty-two participants were randomly assigned to the traditional MS or DR paradigm. In Experiment 2, we explored the effects of different types of death management (i.e., “sorrow management” and “terror management”) on external defenses (i.e., worldview defense and extrinsic life goal structure) and internal growth (i.e., intrinsic life goal structure). Altogether 152 participants were randomly assigned to the traditional MS paradigm, DR paradigm, or control condition. The mediating role of emotions (fear, sadness) was also explored. In Experiment 3, we explored further the effects of terror management and sorrow management on external defenses (i.e., self-esteem striving and extrinsic life goal structure) and internal growth (i.e., intrinsic life goal structure). Altogether 182 participants were randomly assigned to the traditional MS paradigm or control condition with sad or fearful background music.

In Experiment 1, in regard to self-reported emotions, the increase in sadness and fear were the primary emotional reactions after MS, t (81) = 10.39, p < 0.001, Cohen’s d = 1.15; t (81) = 8.20, p < 0.001, Cohen’s d = 0.91, respectively. The absolute values of Cohen’s d for MS effects on other basic emotions (i.e., joy, disgust, anger, surprise) were all smaller than 0.49. Additionally, the paradigm (traditional MS vs. DR) × emotion category (sadness vs. fear) × period (pre-test vs. post-test) interaction was significant on emotion value, F (1, 79) = 10.84, p = 0.001, η_p² = 0.12. People in the DR paradigm (M = 2.24, SD = 1.53) reported more increased sadness than in the traditional MS paradigm (M = 1.03, SD = 1.03), p < 0.001, 95% CI = [1.71]. However, there was no significant difference in the increased fear between the traditional MS paradigm (M = 1.00, SD = 1.01) and DR paradigm (M = 1.17, SD = 1.36), p = 0.702, 95% CI = [−0.42, 0.62]. Moreover, sadness is a more prevalent feeling than fear among participants in the traditional MS or DR paradigm based on their choice between sadness or fear as the dominant emotion, χ² (1, 82) = 47.82，p < 0.001. For physiological response, the skin conductance response (SCR) was significantly higher in the arousal period (M = 1.72, SD = 0.95) than in the baseline period (M = 1.06, SD = 0.72), t (81) =11.47, p < 0.001, Cohen’s d = 1.28; the heart rate (HR) was significantly higher in the arousal period (M = 77.71, SD = 12.64) than in the baseline period (M = 73.48, SD = 12.53), t (81) = 3.68, p < 0.001, Cohen’s d = 0.41. However, there was no significant difference in the changed SCR or HR (arousal period minus baseline period) between the two paradigms, ps > 0.05. In terms of facial expressions, sadness and fear were significantly different between the answering period (i.e., answering about the death-related questions) and thinking period (i.e., thinking about the death-related questions), F (1, 80) = 48.32, p < 0.001, η_p² = 0.38. Specifically, sadness during the thinking period (M = 4.88, SD = 5.19) was significantly greater than during the answering period (M = 3.00, SD = 4.59), p < 0.001, 95% CI = [0.80]; in contrast, fear was significantly less during the thinking period (M = 3.06, SD = 6.15) than during the answering period (M = 4.02, SD = 6.78), p < 0.001, 95% CI = [−0.75, −0.26].

In Experiment 2, there was no significant difference between the two paradigms on external defenses (i.e., worldview defense, extrinsic life goal structure) or internal growth (i.e., intrinsic life goal structure), ps > 0.05. Nevertheless, the self-reported fear significantly mediated the effect of MS on external defense (i.e., extrinsic life goal structure) (see Figure 1), B = 0.23, SE = 0.08, 95% CI = [0.39].

In Experiment 3, the MS effects on internal growth (i.e., intrinsic life goal structure) differed significantly under different background music, F (1, 177) = 4.00, p = 0.047, η_p² = 0.02. Specifically, participants in the traditional MS paradigm (M = 29.31, SD = 3.22) attached more importance to intrinsic life goal structure than participants in the control group (M = 27.79, SD = 3.11) when the background music was sad, p = 0.008, 95% CI = [1.94]. However, the MS effect on the intrinsic life goal structure was insignificant when the background music was fearful, p > 0.05. Besides, the type of background music (i.e., sad, fearful) functioned as a moderator in the relationship between MS and intrinsic life goal structure (see Figure 2), B = 0. 56，t = 2.02，p = 0.045. Specifically, MS prompted participants to place more emphasis on intrinsic life goal structure under the sad background music, simple slope = 0. 52，t = 2.62，p = 0. 009，95% CI = [0.91]; but not under the fearful background music, simple slope = −0.04，t = −0.20，p = 0.840，95% CI = [−0.43, 0.35].

In conclusion, these results indicated that MS elicited both sadness and fear, and sadness was more prevalent among Chinese participants. Moreover, death reflection elicited more sadness than traditional mortality salience. Fear functioned as a mediator in the relationship between MS and external defense. Finally, MS could promote internal growth when sadness (but not fear) is the dominant emotion. The current study deepens the understanding of the role of emotions (especially sadness) in MS’s effect on life goal structure, expanding the existing Terror Management Theory. The present study also provides preliminary empirical evidence for the SMT, informing the practice of psychological rescue in epidemics and other social disasters.

Key words: Terror Management Theory, Sorrow Management Theory, sadness, death reflection, life goal structure