DOI: https://doi.org/10.21203/rs.3.rs-2627633/v1
Objective
This study examined the mediating roles of empathy and self-compassion in the relationship between parental attachment, prosocial behavior, and antisocial behavior among Chinese adolescents after the Jiuzhaigou earthquake.
Methods
A total of 411 adolescents (M = 14.73, SD = 0.91) were assessed 16 months after the Jiuzhaigou earthquake using self-report questionnaires of trauma exposure, parental attachment, empathy, self-compassion, prosocial behavior, and antisocial behavior.
Results
The parental attachment had a positive effect on prosocial behavior via empathy and positive self-compassion, as well as via an indirect path from empathy to positive self-compassion. Parental attachment had a negative effect on antisocial behavior via empathy and negative self-compassion, and had an indirect path from empathy to negative self-compassion.
Conclusions
These findings indicate that enhancing parental attachment may promote prosocial behavior and reduce antisocial behavior of post-earthquake adolescents. Empathy and self-compassion play significant mediating roles between parental attachment, prosocial behavior, and antisocial behavior.
1. The present study examined the mediating roles of empathy and self-compassion in the association between parental attachment, prosocial behavior, and antisocial behavior among Chinese adolescents after an earthquake.
2. Parental attachment can promote prosocial behavior and prevent antisocial behavior among post-traumatic adolescents.
3. Empathy and different components of self-compassion played significant mediating roles between parental attachment prosocial behavior, and antisocial behavior.
An earthquake is a major natural traumatic event that has a serious negative impact on human behavioral response and mental health (Cisler et al., 2011; Ehring, Razik, & Emmelkamp, 2011; Feder et al., 2013). Researchers have focused on the psychological reactions of earthquake survivors, such as depression disorder, post-traumatic stress disorder, and anxiety disorder (Ehlers & Clark, 2000; Guo et al., 2018; Cheng et al., 2013). However, only a few researchers have focused on the behavioral responses of adolescents who experienced traumatic events in earthquakes. Adolescence is the key period of a human’s lifelong physical and mental development, and the impact of traumatic events on adolescents is enormous and far-reaching (Steinberg, 2005). Therefore, studying the internal mechanism of adolescent behavioral response after traumatic earthquake events is of great significance.
Antisocial Behavior and Prosocial Behavior
When individuals experience traumatic events, they often have negative behavioral reactions, and antisocial behavior is one of the representative negative behavioral reactions (Ford et al., 2012). Antisocial behavior is a pattern of behavior characterized by neglect and violation of the interests of others, such as assault, vandalism, deception, and theft (American Psychiatric Association, 2013). Social information processing model (Holtzworth-Munroe, 1992) points out that when individuals experience serious natural disasters, they often have unreasonable beliefs, such as that the surrounding environment is hostile and threatening. These irrational beliefs can cause individuals to exhibit antisocial behaviors such as aggression and destructive behavior (Taft et al., 2012). Most empirical studies also support the view that children and adolescents may exhibit different antisocial behaviors after a major earthquake disaster (Cheng et al., 2013; Wang et al., 2020).
Meanwhile, Vollhardt and Staub (2011) found that after an earthquake, individuals are more likely to actively participate in prosocial behaviors such as rescuing, donating money, and providing volunteer services. These prosocial behaviors have positive and important significance for the mutual aid reconstruction after an earthquake. Vollhardt (2009) described this phenomenon as the altruistic behavior that comes from experiencing suffering.
Prosocial behavior and antisocial behavior often coexist in groups that have experienced catastrophic traumatic events (Eron & Huesmann, 1984; Gomà-i-Freixanet, 1995). Therefore, Are there common factors and underlying mechanisms that influence both of these behavior patterns? This study includes prosocial and antisocial behaviors to provide new insights for psychological intervention in post-earthquake adolescent behavioral patterns.
Parental Attachment affects Antisocial Behavior and Prosocial Behavior
For adolescents, the family environment can promote the development of their social function and shape their behavioral pattern (Hirschi, 1969). Parental attachment, which is defined as the emotional bond between adolescents and their parents (Bowlby, 1969), is an important part of the family environment. It can provide supportive resources for adolescents to cope with the stress caused by traumatic events (Tian et al., 2018). Moreover, parental attachment may play an important role in the prosocial behaviors and antisocial behaviors of individuals who have experienced catastrophic traumatic events.
In terms of the relationship between parental attachment and prosocial behavior, attachment theory (Bowlby, 1969) suggested that individuals are born with caregiving systems. These systems can guide people to pay more attention to the feelings and needs of others and can encourage individuals to provide protection and support to those in need (Mikulincer & Shaver, 2007; Mikulincer et al., 2005). High levels of parental attachment sets the important psychological foundation for a caregiving system that enables individuals to feel safe in providing help without worrying excessively about the internal resources consumed. Hence, High levels of parental attachment promotes individuals’ prosocial behavior (Mikulincer & Shaver, 2007; Richman et al., 2015; Wang et al., 2017).
Previous theories and studies have provided evidence for the relationship between attachment and antisocial behavior. According to social control theory (Hirschi, 1969), antisocial behavior is low in families with strong affective ties, because adolescents who are strongly attached to their parents are more likely to care about the normative expectations of their parents, which are born against antisocial behavior. Existing empirical studies suggest that parental attachment can significantly reduce the level of antisocial behavior in adolescents (Arbona & Power, 2003; Thompson & Gullone, 2008; Bekker et al., 2007).
Although previous studies have generally shown that parental attachment as an environmental factor can influence adolescent behavior patterns, only a few studies have explored the potential mechanisms of parental attachment on prosocial and antisocial behaviors in the context of natural disasters. In previous studies, one potential mechanism of particular interest in understanding the links between parental attachment and behaviors is empathy (Carlo et al., 2012).
Mediating Role of Empathy
As a personality trait, empathy is the ability to accurately feel the feelings of others and understand the meaning of those feelings (Kalisch, 1973). Regarding the association between parental attachment and empathy, attachment theory provides evidence that empathic responses to self and others are based on positive internal working models (Stern & Cassidy, 2018). Individuals with high levels of parental attachment, that is, with positive internal working model about themselves and their environment, perceived others with greater understanding and acceptance. This ability enables them to pay attention to the feelings and needs of others and respond empathically when they feel the suffering of others (Kestenbaum, 1989; Bretherton & Munholland, 2008). Research also supports the relationship between parental attachment and empathy (Schoeps et al., 2020; Thompson & Gullone, 2008).
Furthermore, empathy has been indicated as an important predictor of prosocial behaviors and antisocial behaviors (McGinley & Carlo, 2012). According to the empathy altruism hypothesis proposed by Batson (1987), when individuals witness others’ misfortune, a high level of empathy will cause them to feel compassion and encourage them to implement prosocial behaviors. In a study of adolescents who had experienced an earthquake, empathy was found to have a positive effect on prosocial behavior (Wang & Wu, 2020). However, the relationship between empathy and antisocial behavior from previous studies is not consistent. Some scholars have noted, for example, that feelings of concern for others would inhibit intention to hurt others. Thus, empathy may be expected to negatively predict antisocial behaviors (Eisenberg & Fabes, 1998; Carlo et al., 2012). However, some researchers have suggested that individuals with high empathic abilities are more likely to experience negative emotions when they perceive others’ distress. These negative emotions, if not properly regulated, will increase the occurrence of antisocial behavior (Gill & Calkins, 2003). Therefore, the effect of empathy on antisocial behavior needs to be further explored in trauma samples.
In addition to the personality factor of empathy, what other psychological processes play a role in the relationship between parental attachment, prosocial behavior, and antisocial behavior? Gilbert's theory of emotional regulation system suggests that early parental attachment can cultivate individuals' level of self-compassion (Gilbert, 2014), and the ability of self-compassion will further affect individuals' psychological and behavioral outcomes (Pietromonaco & Barrett, 2000; Jiang et al., 2017). Therefore, self-compassion may work as an underlying mechanism that plays a role between parental attachment and post-traumatic behavioral responses.
Mediating Role of Self-compassion
Self-compassion is defined as an individual’s tendency to understand, forgive, and not judge themselves in the face of suffering (Neff, 2003a). Conceptually, self-kindness, common humanity and mindfulness are the positive components of self-compassion, representing an individual's tolerance and understanding of himself, calmness towards suffering, and rational and clear attitude towards the current predicament. While the three dimensions of self-judgment, isolation and over-identification are negative components, which are a tendency to be extremely harsh to self and excessively exaggerated to pain (Neff, 2003a). Self-compassion is commonly measured as the sum of all items of the Self-compassion Scale (SCS; Neff, 2003b) after reverse scoring the items that assessed negative components. However, reliance on the SCS total score as an indicator of self-compassion is debatable. The positive and negative components of self-compassion are relatively independent in structure and have different effects on an individual’s mental health (Brenner et al., 2017; Muris & Petrocchi, 2016; Muris et al., 2018). Furthermore, Mackintosh et al. (2018) pointed out that future studies on the relationship between attachment, self-compassion, and psychopathology must account for the different dimensions of self-compassion. Therefore, we divide self-compassion into positive self-compassion and negative self-compassion in this study.
The relationship between parental attachment and self-compassion has been supported by earlier theories. Attachment theory (Bowlby, 1969, 1979) suggested that early life experiences of attachment images shape key internal working models of the self. Individuals with positive attachment are more likely to form positive internal working models of self, because they fully feel the security and support provided by attachment figures. This security enables individuals to form values worthy of being cared about and to concern and love themselves when they are suffering (Mikulincer & Shaver, 2007). Thus, high quality parental attachment may enable individuals to employ strategies of positive self-compassion when experiencing negative events (Homan, 2018; Jiang et al., 2017). In terms of the relationship between attachment and negative self-compassion, those with a higher level of insecure attachment are likely to develop a negative view of self (Pietromonaco et al., 2000) and to be self-critical in suffering (Cantazaro & Wei, 2010). Naturally, self-criticism and blame are important manifestations of negative self-compassion. In an empirical study of adults, low levels of parental attachment was found to be a significantly positive predictor of negative self-compassion (Brophy et al., 2020).
As an emotional regulation strategy, self-compassion can further influence individuals’ post-traumatic behavioral responses. In terms of the relationship between positive self-compassion and prosocial behavior, a person with positive self-compassion is expected to be more likely to recognize when another is suffering from negative events. This ability is an important step in showing prosocial behavior (Atkins & Parker, 2012). Researchers suggest that building the capacity for caring and accepting oneself in suffering can help extend compassion to others. Such capacity makes individuals more likely to help others when they witness their suffering (Hofmann, et al., 2011). Thus, positive self-compassion may facilitate prosocial behavior (Longe, 2009; Crocker & Canevello, 2008; Liu et al., 2021). However, individuals with high negative self-compassion may focus on their own painful emotions and interpersonal estrangement from others, which may lead to social withdrawal and loneliness, neither of which are conducive to engaging in prosocial behaviors (Charmaz, 1980; Kahn, 1965).
Self-compassion and antisocial behavior are also correlated. In the face of setbacks and failures, positive self-compassion can reduce the risk of antisocial behavior by alleviating negative emotional reactions such as anger and impulsivity (Leary, 2007; Neff & Beretvas, 2013). In contrast, people with higher levels of negative self-compassion tend to make excessive demands to others as well as being hard on themselves (Kahn, 1965). If the demands and expectations of others are not met, feelings of dissatisfaction, anger and hostility will arise, which may lead to aggressive behavior (Milrod, 1972; Wilson, 1985).
Although empathy and self-compassion are potential mediators of the relation between parental attachment, prosocial behavior, and antisocial behavior, these two mediators have a combined relationship. Theoretically, when a person with a high level of empathy witnesses someone else’s misfortune, his or her own painful emotions may be triggered (Loggia et al., 2008). Moreover, self-compassion naturally arises from the affective component of empathy when one desires to alleviate the negative emotions through self-regulation (Steffen & Masters, 2005). Existing studies have demonstrated a positive association between empathy and positive self-compassion (Marshall et al., 2020; Rachel et al., 2018). Although a few studies have examined the relationship between empathy and negative self-compassion, evidence shows an association between empathy and variables with concepts similar to negative self-compassion. Heym et al. (2019) mentioned in their study that individuals with high levels of empathy may be at risk of over-identification due to excessive emotional involvement when witnessing others’ misfortune. However, in a study of medical workers, empathy was found to be significantly negatively correlated with self-judgment (Beaumont et al., 2016). Therefore, the possible effects of empathy on negative self-compassion have not been consistent in previous studies. The relationship between these two variables needs to be explored in future studies.
Present study
Empathy and self-compassion can theoretically mediate the relationship of parental attachment with prosocial/antisocial behavior. However, to our knowledge, no empirical study has simultaneously examined the mediating roles of empathy, positive self-compassion, and negative self-compassion in the relationship between parental attachment and prosocial/antisocial behavior. Therefore, the main objective of the current study is to examine how adolescents’ parental attachment after the Jiuzhaigou earthquake predicted prosocial/antisocial behavior by including empathy, positive self-compassion, and negative compassion as mediating variables.
Participants and procedures
One and a half years after the Jiuzhaigou earthquake, we contacted local education authorities in the most severely affected county. We informed them of the aims and methods of this study, and offered to provide psychological services when they were required. With the help of these authorities, we obtained approval from two middle schools to recruit students for our research, and then randomly selected several classes from each school. All the selected students agreed to participate; they comprised 411 adolescent earthquake survivors. The mean age was 14.73 (SD = 0.91), ranging from 12.0 to 17.0 years. Of the 411 participants, 225 (54.7%) were female, 186 (45.3%) were male, and one did not report gender.
Participants came from two middle schools in Jiuzhaigou County in Sichuan Province. Both schools were affected by the earthquake. The study protocol was approved by the research ethics committee of Beijing Normal University (No. 202003190026). Before the formal investigation, we solicited the approval of the principals and teachers of the two middle schools. Additionally, we obtained the signed consent forms from students and their guardians on a voluntary basis. All questionnaires were recalled immediately after completion.
Measures
Trauma exposure. The trauma exposure questionnaire revised by Qi et al. (2020) was adopted to measure the severity of Chinese adolescent survivors’ traumatic experiences. This scale measured seven items, including whether participants witnessed harm or death and whether participants were notified afterwards of the harm or death of others. The scoring standard of this scale is a two-point scoring system, with each item having a score range of 0 (no) to 1 (yes). The sum of each participant’s scores considered an indicator of trauma exposure, and the scoring range of this questionnaire was from 0 to 7. The internal reliability of the questionnaire used is acceptable, and the Cronbach’s α of the questionnaire is 0.68.
Parental attachment. In this study, the Inventory of Parent and Peer Attachment (Armsden & Greenberg, 1989) was used to assess parental attachment of adolescents. This questionnaire provides an overall score for attachment security and scores on the three subscales: trust, communication, and alienation. In each sub-scale, the item format was a five-point Likert scale (1 = almost always or always true, 2 = often true, 3 = sometimes true, 4 = rarely true, and 5 = almost never or never true). In the present sample, the sub-scales of parental attachment showed good internal consistency reliability, and the Cronbach’s α of the questionnaire is 0.87.
Empathy. In this study, the 28-item self-report questionnaire Interpersonal Reactivity Index (IRI) (Davis, 1980) was used to assess participants’ empathy. This questionnaire is composed of four subscales (7 items each): perspective taking, fantasy scale, empathetic concern, and personal distress. Items range in score from 1 (does not describe me well) to 5 (describes me very well). According to Siu and Shek (2005), in the IRI, the items most representative of empathy are mainly reflected in the two dimensions of perspective taking and empathetic concern. On this basis, the two dimensions of viewpoint selection and empathic concern were selected in this study. The reliability of the scale was acceptable, and the Cronbach’s α of the questionnaire is 0.74.
Self-compassion. In this study, the 26-item SCS compiled by Neff (2003a) was used to measure participants’ self-compassion. This questionnaire was composed of six subscales, which were divided into positive self-compassion and negative self-compassion. Positive self-compassion includes three different dimensions, namely, common humanity, self-kindness, and mindfulness. Negative self-compassion includes three different dimensions, namely, isolation, self-judgment, and over-identification. The five-point Likert scale used in this scale had items that range in score from 1 to 5. Moreover, the applicability of this scale has been verified in post-traumatic Chinese adolescents (Liu, Wang, Wu, 2020). The internal reliability of PSC (Cronbach’s α = 0.70) and NSC (Cronbach’s α = 0.80) was good.
Antisocial behavior. To assess antisocial behavior, participants completed the subscale of the Achenbach’s Youth Self-report version of the Child Behavior Checklist (Achenbach & Edelbrock, 1987). This scale includes nine items, and a three-point Likert Scale is used in this scale. The items range in score from 0 (none) to 2 (frequently). This scale shows good test–retest reliability and criterion validity in Chinese adolescents (Lan, Marci, & Moscardino, 2019). The internal reliability of the scale was acceptable (Cronbach’s α = 0.73).
Prosocial behavior. In this study, the Prosocial Behaviors Scale was used to measure prosocial behaviors in Chinese adolescents. The scale is composed of 15 items, and the scale was specially developed for Chinese adolescents (Zhang & Kou, 2008). This scale is composed of four different dimensions, which are altruism, commonwealth-rule, interpersonal relationships, and personality traits. The Likert scale used in this study had items that range in score from 1 to 7. In the present sample, the Cronbach’s alpha index of internal consistency reliability for the scale was good (Cronbach’s α = 0.91).
Data analysis
In this study, SPSS (Windows 24.0) and AMOS (22.0) software were used to analyze our data. The missing value for item level was minimal (i.e., the missing value for each individual was not more than 1.5%) and the data were missing at random. so were handled with maximum likelihood estimates (ML) in structural models.
In this study, we used the Chi-square values, comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), and 95% confidence interval (CI) for model fit. According to academic consensus (Hu & Bentler, 1999), when CFI and TLI are greater than or equal to 0.09 and RMSEA is less than or equal to 0.08, the overall model fits well.
Descriptive statistics and correlations among measures
Table 1 presents the correlation analysis among major variables.
[Table 1 near here]
Structural equation model analyses
To test our hypothesis, we used four steps to examine the multiple mediating effects of empathy and self-compassion on the relationship between parental attachment, prosocial behavior, and antisocial behavior. First, we test the measurement model. The measurement model in this study consists of five latent variables, and each latent variable is extracted from the dimensions of each scale. According to the test results, the fitting indexes of the measurement model were acceptable: χ2/df = 2.17, CFI = 0.957, TLI = 0.949, RMSEA (90% CI) = 0.061 (0.048–0.075). The factor loadings of the indicators in the latent variables were significant (p < 0.001).
Second, we established a direct effects model to test the influence of parental attachment on prosocial behavior and antisocial behavior (FIGURE 1). The fitting indexes of the direct effects were acceptable: c2/df = 3.56, CFI = 0.954, TLI = 0.923, RMSEA (90% CI) = 0.079 (0.061−0.098). The results of path analysis showed that parental attachment had a positive effect on prosocial behavior and a negative effect on antisocial behavior.
[FIGURE 1 near here]
After controlling for gender, age, and traumatic exposure, we built a mediating effect model to examine the mediating roles of empathy and self-compassion between parental attachment, prosocial behavior, and antisocial behavior. The fitting indexes of the indirect effects were acceptable: c2/df = 2.09, CFI = 0.962, TLI = 0.947, RMSEA (90% CI) = 0.052 (0.041−0.062). According to the results of path analysis, parental attachment had a negative effect on antisocial behavior, but the predictive effect on prosocial behavior is not significant. Moreover, parental attachment had an indirect and positive effect on prosocial behavior via empathy and positive self-compassion, as well as via an indirect path from empathy to positive self-compassion. Parental attachment had a negative effect on antisocial behavior via empathy and negative self-compassion, but had an indirect and positive path from empathy to negative self-compassion on antisocial behavior.
[FIGURE 2 near here]
Finally, bias-corrected bootstrap tests were performed on the path with significant indirect effect in the path analysis. Table 2 illustrates the results of these tests. The 95% CI of the above mediating effect did not contain 0, and the mediating effect was significant. The indirect effect accounts for 43.6% of the total effect.
[TABLE 2 near here]
This study have investigated the mediating roles of empathy and self-compassion in the relationships between parental attachment and prosocial/antisocial behavior. In the direct effects model, parental attachment has direct positive effects on prosocial behavior, which is consistent with previous studies (Mikulincer & Shaver, 2007; Wang, Sun, Zhao, Lai, & Zhou, 2017; Richman, DeWall, & Wolff, 2015). According to attachment theory (Bowlby, 1969), caregiving systems are an innate ability of human beings to directs one’s attention toward others’ distress and demands rather than exclusively focusing on his/her own vulnerability, provides protection, and support to those in need. The sense of security and supportive resources provided by high levels of parental attachment can promote the formation of adolescents’ caregiving system (Mikulincer et al., 2005), and then improve the level of adolescents’ prosocial behavior. Parental attachment can also play a significant negative predictive role for antisocial behavior, which is consistent with previous studies (Arbona & Power, 2003; Thompson & Gullone, 2008; Bekker et al., 2007). Therefore, high levels of parental attachment can make adolescents care more about what their parents expect of them, thus reducing the likelihood of antisocial behavior.
When considering relationships explained by empathy as a standalone mediator, parental attachment was associated with high levels of empathy, and concomitantly with more prosocial behavior and less antisocial behavior. These results were supported by findings from various prior studies (Schoeps, et al., 2020; Ardenghi et al., 2019; Wang & Wu, 2020; Carlo et al., 2007; Berger et al., 2015). Attachment theory (Bowlby, 1979) proposes that patterns of caregiving behavior shape an adolescent’s internal working model of the self and others, which is the basis of an individual’s ability to empathize. That is, high levels of parental attachment enable adolescents to have a positive internal working model, which leads to a stronger emotional resonance to others around them and a prompt response to others’ needs, which ultimately promotes their ability to empathize (Cozolino, 2006; Siegel, 2001). Adolescents with high levels of empathy were more likely to feel sorrow about others’ suffering. On the one hand, feelings of sorrow for someone else are an important motivation to alleviate others’ distress, and thus, to show more prosocial behavior toward others (Batson, 1991; Batson et al., 1989). On the other hand, individuals who experience high levels of empathy and related processes are presumed to feel responsibility toward others, and as a result are motivated to reduce their distress (Eisenberg & Fabes, 1998). In this study, therefore, adolescents with high levels of empathy showed lower levels of antisocial behavior.
Furthermore, the positive and negative components of self-compassion played an important mediating role in the relationship between parental attachment and prosocial/antisocial behavior. Specifically, parental attachment has a positive effect on prosocial behavior through positive self-compassion, which is consistent with previous studies (Homan, 2018; Jiang, et al., 2017; Longe, 2009; Crocker & Canevello, 2008; Liu et al., 2021). This result suggests that high levels of parental attachment is more likely to form positive internal working models of self, thus enabling adolescents to form values that they are worthy of being loved and cared for (Mikulincer & Shaver, 2012; Shaver et al., 2017). As a result, their level of positive self-compassion is improved. Adolescents who were adept in self-care after the earthquake were more likely to extend their kindness to others (Hofmann et al., 2011), thus exhibiting more prosocial behavior. In addition, parental attachment can negatively predict antisocial behavior by reducing negative self-compassion, which is consistent with results from recent studies (Brophy et al., 2020; Liu et al., 2021). The positive inner working model developed by high-quality parental attachment can promote individuals to view themselves and others around them from a positive perspective. Such adolescents usually have high self-evaluation and good interpersonal relationships (Mikulincer & Shaver, 2012; Shaver et al., 2017); hence, they are less likely to use negative self-compassion as an emotional regulation strategy when experiencing traumatic events. Considering that painful emotions and social isolation induced by negative self-compassion may be risk factors for antisocial behavior (Kahn, 1965; Milrod, 1972; Wilson, 1985; Liu et al., 2020), parental attachment can reduce the incidence of antisocial behavior by alleviating negative self-compassion.
Moreover, parental attachment has a two-mediator indirect effect on prosocial behavior and antisocial behavior through empathy via self-compassion. High quality parental attachment can cultivate the ability to empathize (Kestenbaum, 1989; Bretherton & Munholland, 2008). A high level of empathy can play a role in adolescents’ self-compassion, including positive self-compassion and negative self-compassion. In the aftermath of the earthquake, empathy for others who have also suffered negative events may result in unpleasant emotions, such as sadness and guilt, especially for adolescents (Loggia et al., 2008). Self-compassion naturally arises from the affective component of empathy when one desires to deal with negative emotions through self-regulation (Steffen & Masters, 2005). When the negative emotions caused by empathy bring painful feelings to individuals, the motivation of self-improvement will urge individuals to adopt the emotional regulation strategy of positive self-compassion to care for and treat themselves well (Breines & Chen, 2012), so as to help them alleviate negative psychological reactions. An individual who treats himself kindly and gently is more likely to exhibit prosocial behavior. However, the psychological impact from empathy for the misfortunes of others may trap an individual in pain for a long time, leading to strong feelings of loneliness and self-criticism (Heym et al., 2019). Therefore, empathy may increase people’s levels of negative self-compassion, which, in turn, promotes antisocial behavior. In general, parental attachment can promote adolescents’ prosocial behavior through the chain mediating effect of empathy and positive self-compassion. It can also increase antisocial behavior through the mediating path from empathy to negative self-compassion.
Finally, parental attachment could not influence antisocial behavior through positive self-compassion, and it could not influence prosocial behavior through negative self-compassion. Although this result is inconsistent with the results of previous studies (Neff & Beretvas, 2013; Charmaz, 1980; Kahn, 1965), it echoes the debate about the structure and concept of self-compassion in previous studies. Existing research has shown that the positive and negative components of self-compassion have unique effects on an individual’s psychological and behavioral responses, rather than the opposite (Brenner et al., 2017; Coroiu et al., 2018; Gilbert et al., 2006). Specifically, positive self-compassion has a stronger effect on positive psychological and behavioral outcomes, whereas negative self-compassion can induce more adverse psychological and behavioral responses (Montero-Marin et al., 2018; Muris and Petrocchi, 2016; Muris et al., 2018; Pfattheicher et al., 2017). In this study, although positive self-compassion promoted prosocial behavior, self-care and caring did not mean that antisocial behavior was prevented. In fact, a few scholars believe that positive self-compassion is a form of egoism, which may seek to comfort oneself by damaging the interests of others (Marshall et al., 2020). Similarly, although people with negative self-compassion are more likely to engage in antisocial behavior, it does not mean that it may further block prosocial behavior. Moreover, self-criticism and shame, which are associated with negative self-compassion, may even prompt individuals to exhibit altruistic behavior (Gausel et al., 2016). Therefore, at least in this study, positive and negative self-compassion, as two separate concepts, have different, and not only opposite effects, on the behavioral responses of post-traumatic adolescents.
Several limitations of the current study should be acknowledged. First, all data were collected by self-report scales, which may be susceptible to common method bias. In addition, prosocial behavior and antisocial behavior, for example, are likely to be influenced by the social desirability effect, leading to bias in the results. Future studies should use more evaluation methods to collect data. Second, this study explores the effect of empathy on self-compassion, and this relationship can be traced theoretically. However, the causal relationship between the two seems is not clear in previous studies. Therefore, collecting data on empathy and self-compassion at different times is necessary to examine their causal relationship.
Despite these limitations, this study used longitudinal analysis, which provides meaningful insights into the relationships between parental attachment, empathy, self-compassion, and prosocial and antisocial behaviors in the context of natural disasters. First, this study explores the potential mechanisms of family environmental factors on adolescents’ post-traumatic behavior responses from positive and negative perspectives. The findings not only confirm the applicability of attachment theory to the field of trauma, but also establish a dialectical integration perspective to explore the double-edged role of personality and emotion regulation strategy factors in family relationships and adolescents’ post-traumatic behavioral responses. Specifically, parental attachment can improve adolescents’ empathic ability and further enhance adolescents’ adaptive behavior through the emotional regulation strategy of positive self-compassion. However, empathizing with the misfortune of others may also activate adolescents’ negative self-compassion and lead to adverse behavioral responses. Therefore, psychological intervention workers should pay attention to adolescents’ attachment relationship, and if necessary, carry out family counseling to promote their harmonious relationship with their parents, so as to solve adolescents’ post-traumatic behavior problems. In addition, a dialectical perspective should be adopted in the intervention of post-traumatic adolescent behavioral response. Empathy and self-compassion, for example, may play positive and negative roles and lead to different post-traumatic behavioral responses.
Ethics approval and consent to participate
Informed consent was obtained from participating adolescents and their legal guardians. The study protocol was approved by the research ethics committee of Beijing Normal University (No. 202003190026). All procedures performed in this study involving human participants were in accordance with the ethical standards of Beijing Normal University and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Consent for publication
Not applicable
Availability of data and materials
The datasets generated during and/or analyzed during the current study are available at https://pan.baidu.com/s/1lXQtRrT5zMC7t1kxxh8gKg?pwd=gzl1.
Funding
The current study was supported by “The National Natural Science Foundation of China” (Project No. 32071085) and “The Fundamental Research Funds for the Central Universities of China” (Project No. 2020NTSS02).
Authors’ contributions
A.L. was conceived of the study, performed the statistical analysis and interpretation, and drafted the manuscript; B.X. and W.W. participated in the design and helped to revise the manuscript; X.W. was involved in the conceptualization, ethical approval, interpretation and writing. All authors read and approved the final manuscript.
Competing interests
The authors have no conflicts of interest to declare that are relevant to the content of this article.
Acknowledgements
We would like to thank the staff and administrators at our participating school sites. We are especially indebted to the adolescents and their parents and teachers whose participation made this research possible.
TABLE 1 Means, Standard Deviations, and Correlations of Main Variables |
||||||||||
Variables M±SD |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
|
1 Gendera |
—— |
1 |
|
|
|
|
|
|
|
|
2 Age |
14.73±0.91 |
-0.15** |
1 |
|
|
|
|
|
|
|
3 Traumatic Exposure |
1.42±1.58 |
0.04 |
0.05 |
1 |
|
|
|
|
|
|
4 PA |
83.30±14.91 |
0.05 |
-0.01 |
-0.08 |
1 |
|
|
|
|
|
5 Empathy |
46.89±6.42 |
0.26*** |
-0.05 |
-0.01 |
0.23*** |
1 |
|
|
|
|
6 PSC |
38.93±7.01 |
-0.01 |
-0.03 |
0.04 |
0.25*** |
0.37*** |
1 |
|
|
|
7 NSC |
37.91±7.70 |
0.08 |
-0.16** |
0.05 |
-0.12* |
0.19*** |
0.29*** |
1 |
|
|
8 PB |
73.98±15.54 |
0.14** |
-0.03 |
-0.01 |
0.24*** |
0.46*** |
0.36*** |
0.07 |
1 |
|
9 AB |
2.78±2.33 |
-0.31*** |
0.08 |
0.21*** |
-0.22*** |
-0.22*** |
-0.06 |
0.17*** |
-0.28*** |
1 |
Note. Codeda: 1 = male, 2 = female; PA = parental attachment; PSC = positive self-compassion; NSC = negative self-compassion; PB = prosocial behavior; AB = antisocial behavior.
* p < 0.05, ** p < 0.01, ***p < 0.001.
TABLE 2
Bias-Corrected Bootstrap Tests of Mediating Effects
Path |
Standardized β |
Standardized 95% CI |
|
Low High |
|||
PA-PSC-PB |
0.046* |
0.003 |
0.098 |
PA-NSC-AB |
0.031* |
-0.016 |
-0.079 |
PA-Empathy-PB |
0.125*** |
0.049 |
0.232 |
PA-Empathy-AB |
-0.083** |
-0.057 |
-0.038 |
PA-Empathy-PSC-PB |
0.036* |
0.002 |
0.082 |
PA-Empathy-NSC-AB |
0.012* |
0.002 |
0.040 |
Note: PA = parental attachment; PSC = positive self-compassion; NSC = negative self-compassion; PB = prosocial behavior; AB = antisocial behavior.
*p < 0.05. **p < 0.01. ***p < 0.001.