In this study, we broadened previous research by exploring the relationship between shame and dual harm. Furthermore, we analyzed which type of dual harm can reduce shame more, thus providing a basis for the second path of the cognitive-emotional model of dual harm. In addition, this study included cognitive flexibility, a cognitive factor that plays an important role in emotional development, to explore the possible mitigating effect of cognitive flexibility. We proposed three hypotheses:
H1 Shame triggers dual harm.
H2 Self-injury is more effective at alleviating shame than aggression.
H3 Cognitive flexibility alleviate the impact of shame on dual harm.
Experiment 1: The Influence of Shame on Dual Harm
Method
Participants
Using the Adolescent Self-Injury Questionnaire and the Reactive-Proactive Aggression Questionnaire. Eighty-two Chinese middle school student, aged 12–16 years (34 males: M = 13.97, SD = 0.80; 48 females: M = 14.02, SD = 0.91) are recruited from a middle school in Hebei, China. All participants are right-handed, physically and mentally healthy, with normal or corrected vision. Ethical procedures are followed, and both participants and their guardians are informed and signed an Informed Consent Form. Participants receive appropriate compensation after the experiment.
Design
This experiment use a 2 (shame: priming, control) × 2 (time: pre-test、post-test) two-factor mixed experimental design. The between-group variable is shame, the within-group variable is time. And the dependent variables are the scores of the two dimensions of dual harm (self-injury and aggression), namely, the length of time(s) the subjects put their hands in cold water and the level of chili powder selected by the subjects in the hot sauce experiment.
Materials
The Adolescent Self-injury Behavior Questionnaire.
The scale includes 19 questions and consists of two parts28. The total score is the sum of the product of the number of self-injury behaviors and the degree of harm. A score of 0 indicates no self-harm behavior, otherwise it indicates self-harm behavior. The higher the score, the more serious the self-harm behavior. This study use this questionnaire to measure the self-harm situation of the subjects "in the past year". The Cronbach’s α is 0.817.
The Reactive-Proactive Aggression Questionnaire.
The scale contains 6 questions, divided into two dimensions: reactive aggression and proactive aggression29. A 5-point scoring system is used. Higher scores indicate more severe aggressive behavior. In this study, the Cronbach’s αis 0.771.
Shame priming.
The typical shame events are used to induce shame emotions in the subjects30. The shame event described "you" being caught by the bookstore owner for stealing books from the bookstore, which is about 300 words. After reading, the subjects are asked to self-assess their corresponding emotions, with "1" indicating "not at all" and "7" indicating "very shameful".
Dual harm task.
The hot sauce experiment and the cold pressor experiment are used 31. The procedure of the hot sauce experiment is as follows: the subject first receive a completed taste preference questionnaire. The experimenter tell the subject that this is completed by subject A (virtual subject) in the next room, and as shown in the questionnaire, subject A do not like spicy food. Then the subject is told to choose one from 5 levels of chili powder ("1 = not at all" to "5 = add a lot") and add it to the hot sauce recipe of subject A. The selection process is kept confidential. The specific procedure of the cold pressor experiment is as follows: the subject is told to spread out the non-dominant hand, put the hand into a circulating cold water basin at a constant temperature of 2°C for as long as possible, submerge the water to the wrist, and tell the subject that the test will last for 5 minutes, but the subject is allowed to remove the hand at any time.
Procedure
The subjects sat independently in a quiet laboratory. The experimenter introduce the experiment and ask the subjects to read and sign the informed consent form. Then the first step is carried out. The subjects are divided into two groups, one for the hot sauce experiment and the other for the cold pressor test. They rate their degree of shame on a 7-level scale. In the second step, the subjects complete the shame priming task and answer the retrospective question. Then, the first step is repeated, and the hot sauce experiment and the cold pressor test are carried out respectively and the shame questions are answered. Finally, the experimenter give the corresponding experimental remuneration.
Results
Manipulation Check
The results of the independent sample t-test showed that compared with the control group, the priming group reported more shame after completing the shame priming task, t(80) = 23.24, p < 0.001, d = 0.62. This shows that shame priming is effective.
Dual Harm Task.
The results of repeated measures analysis ANOVA (see Table 1) shows that, on self-harm, the main effect of shame is significant, F (1,39) = 6.49, p < 0.05; η2 = 0.14, and the main effect of time is significant, F (1, 39) = 20.41, p < 0.001; η2 = 0.34. The interaction effect between shame and time is significant, F (1,39) = 5.72, p < 0.05; η2 = 0.13. The simple effect analysis is shown in Fig. 1. In terms of aggressive behavior, the main effect of shame is significant, F (1,39) = 4.32, p < 0.05; η2 = 0.10, and the main effect of time is significant, F (1,39) = 28.74, p < 0.001; η2 = 0.42. The interaction effect between shame and time is significant, F (1,39) = 15.24, p < 0.001; η2 = 0.28. The simple effect analysis is shown in Fig. 2.
Table 1
Scores of the two groups of subjects on different types of dual harm(M ± SD)
Shame | Dual harm |
Self-injury | Aggression |
Pre-test | Post-test | Pre-test | Post-test |
Priming | 104.71 ± 42.80 | 151.94 ± 32.07 | 2.29 ± 0.69 | 3.35 ± 0.86 |
Control | 93.95 ± 36.97 | 107.79 ± 47.43 | 2.29 ± 0.81 | 2.83 ± 0.72 |
Shame Task.
In order to further explore which type of dual injury can better reduce shame, an independent sample t test is conducted on the difference in shame scores between the pre- and post-tests of the initiation-self-injury group and the initiation-attack group, see Table 2. The results showed that the two The difference in group score difference is significant, t(32) = 4.29, p < 0.001, d = 0.80, indicating that self-injury can alleviate shame better than aggressive behavior.
Table 2
Shame scores of subjects with different types of dual injury over time(M ± SD)
Dual harm | Shame-priming group |
Pre-test | Post-test |
Self-injury | 2.76 ± 1.83 | 1.81 ± 0.70 |
Aggressive | 2.78 ± 1.85 | 2.07 ± 1.06 |
Experiment 2: The Influence of Shame on Dual Harm
Method
Participants
Using the Adolescent Self-Injury Questionnaire and the Reactive-Proactive Aggression Questionnaire. Eighty-two Chinese middle school student, aged 12–16 years (44 males: M = 14.03, SD = 0.87; 48 females: M = 14.00, SD = 0.83) are recruited from a middle school in Hebei, China. All participants are right-handed, physically and mentally healthy, with normal or corrected vision. Ethical procedures are followed, and both participants and their guardians are informed and signed an Informed Consent Form. Participants receive appropriate compensation after the experiment.
Design
This experiment use a 2 (shame: priming, control) × 2 (cognitive flexibility: high, low) two-factor between-subjects design. The independent variables are shame and dual harm type. The dependent variable is dual harm score, that is, the summed mean of the ratings of how long the subjects kept their hands in the cold water and the ratings of chili powder selected in the hot sauce experiment.
Materials
Shame priming and dual harm experiments are the same as Experiment 1.
Cognitive flexibility.
The cognitive flexibility task uses the Wisconsin card sorting task(WCST)32. The task is to sort cards according to color, shape and bibliography, with an interval of 500 milliseconds between cards. After each sorting, the computer automatically gives correct or incorrect feedback, and the subjects switch between the classification dimensions. The total number of correct classifications of the subjects is used as the standard for distinguishing high and low cognitive flexibility.
Procedure
The subjects sat independently in a quiet laboratory. The experimenter introduce the experimental situation and aske the subjects to read and sign the informed consent form. Then the first step is carried out. The subjects complete the WCST and the shame priming task and answer the retrospective question. Then,the subjects complete the hot sauce experiment and the cold pressor experiment tasks, and rate the degree of shame they experienced on a 7-level scale. Finally, the experimenter give the corresponding experimental compensation.
Results
Manipulation Check
The results of the independent sample t-test showed that compared with the control group, the priming group reported more shame after completing the shame priming task, t(90)=27.68, p < 0.001, d = 0.60. This shows that shame priming is effective.
Dual Harm Task.
A paired sample t test is conducted on the results of different groups, see Table 3. The results show that there is no significant difference in the pre-test between each group. An independent t-test is conducted on the post-tests of each group and find that only the dual harm behavior of the participants in the priming-cognitive inflexibility group has significant differences in the pre- and post-tests.
Table 3
Paired samples t-test for dual harm behaviors in different groups (N = 92)
Cognitive flexibility | Dual harm |
Shame-control group | Shame-priming group |
High | 2.05 ± 0.65 | 2.35 ± 0.52 |
Low | 2.00 ± 0.85 | 2.97 ± 0.54 |
Analyzing the experimental data of different groups in sequence, it is found that the main effect of cognitive flexibility is not significant, F (1, 88) = 2.83, p = 0.09; η2 = 0.03, that is, subjects with different cognitive flexibility are in dual situations. There are no significant differences in harmful levels. The interaction effect between shame and cognitive flexibility is significant, F (1,88) = 6.26, p < 0.05; η2 = 0.07. The results of simple effects analysis show that under the shame priming condition, see Fig. 3, the main effect of cognitive flexibility is significant, F (1,88) = 4.33, p < 0.05; η2 = 0.05, that is, the shame priming group subjects engage in less dual harm behavior under cognitive flexibility.