The Association between Attachment and Moral Emotions: A Multi-Level Meta-Analysis
Masterthesis Forensic Child and Youth Care Sciences Graduate School of Child Development and Education
University of Amsterdam W. M. Oward
10416722 Supervisors: Prof. dr. F.W. Schalkwijk & Prof. dr. G.J.J.M. Stams
Abstract
This multilevel meta-analysis was performed to examine the overall relation between
attachment and moral emotions (i.e., empathy, shame, and guilt). In total, 20 studies with 22 independent samples, reporting on 142 effect sizes, were included. The total sample consisted of N = 7,457 participants. Results showed a small and significant association (r = .153) between attachment and moral emotions, which may be a substantial underestimation of the true relation. Moderator analyses showed a significant association for categorical attachment measures (r = .241), and a non-significant association for continuous attachment measures (r = .063). The discussion focuses on the relation between attachment and delinquency, which seems to be mediated by moral emotions rather than moral cognition. Special attention should be paid to shame, because coping with shame seems more important than the degree to which children or adolescents experience shame. The results of this study add to the existing
knowledge on moral development, and may thus increase our understanding of the relation between attachment and delinquency.
Table of Contents 1. Introduction ... 4 1.1. Attachment ... 6 1.2. Moral emotions ... 8 1.2.1. Empathy ... 9 1.2.2. Shame ... 10 1.2.3. Guilt ... 11
1.3. The association between attachment and moral emotions ... 12
1.4. The present meta-analyses ... 13
2. Method ... 14
2.1. Selection of studies ... 14
2.2. Coding and moderators ... 16
2.3. Calculation and analysis of effect sizes ... 17
2.4. Publication bias ... 18
3. Results ... 19
3.1. Overall effect size ... 19
3.2. Publication bias and Moderator analysis ... 19
4. Discussion ... 22
4.1. Limitations, strengths and recommendations ... 25
5. Conclusion ... 28
References ... 29
1. Introduction
Joshua is a 9-year-old boy in an out-of-home placement facility, who is caught stealing sweets from the other children. To teach him a lesson the caregiver decides to take his sweets and divide them amongst the other children. When Joshua is asked if he understands why his sweets are divided, he answers that it is because he was caught stealing. Shortly after, he goes upstairs and tears apart a drawing he had made and given to the caregiver as a payback for dividing his sweets. In this vignette Joshua feels ashamed, because the caregiver shamed him. He restores his sense of self by becoming angry at the person who shamed him. Shame, which can be considered as a self-conscious emotion, has to be regulated as it touches on one’s sense of self or identity (Tangney et al., 2007).
The example clearly shows how Joshua rebalances his esteem. Balancing self-esteem and feelings is important for everyone. Self-self-esteem regulation is taken care of by the conscience, and is closely related to the development of a person's identity (Schalkwijk, 2015; Schalkwijk et al., 2016; Schalkwijk et al., in press). The development of conscience begins in early childhood when the child starts developing psychological competencies that are needed for a proper development of the conscience (Schalkwijk, 2015; Schalkwijk et al., in press, Thompson, 2007).
The earliest form of conscience, affective empathy, develops in the first two years of life. Then, self-conscious emotions, such as shame, pride, and guilt come to the fore, followed by the development of cognitive empathy and moral judgement (Schalkwijk et al., 2016). The development of these psychological faculties affects the way young children construct, and act upon, internalized standards of behavior (Thompson, 2007; Thompson et al., 2006). The intersection of these morally relevant psychological faculties makes young children develop views of themselves in relation to others and the broader social world (Thompson, 2007), and ensures that self-esteem can be balanced (Tracy et al., 2007).
In general, conscience functions in the background of consciousness that, usually in an unconscious way, balances the stability of self-esteem. When there are serious shortcomings in these capabilities in early childhood, a solid foundation for the development of conscience is lacking, and as previously mentioned, it is very important to balance the stability of self-esteem and feelings (Lichtenberg et al., 2011; Schalkwijk, 2015). If the conscience cannot regulate the disturbing factors when the balance is threatened, self-conscience emotions emerge in consciousness, and lead to self-defense through the use of defense mechanisms. On the other hand, when these capabilities are well developed, the child is able to experience self-conscious emotions and knows how to regulate them (Schalkwijk, 2015; Schalkwijk et al., in press; Tangney et al., 2007). The child becomes increasingly capable of experiencing
empathy, self-conscious emotions as guilt, pride or shame, and starts making moral
considerations. These three factors form the basis for the psychological domain of conscience. In early childhood, the conceptual basis for moral development becomes established and the growth of the conscience, together with moral development, forms moral behavior
(Thompson, 2007; Thompson et al., 2006).
Besides self-regulation, conscience also has a function in the regulation of social relationships, so it related with attachment (Stern & Cassidy, 2018; Stilwel et al., 1997). When a person has internalized how relationships should be, conscience develops through empathy, self-conscious emotions, and morality (Schalkwijk et al., 2016). From this point, young children are interested in the feelings, thoughts, and intentions of others instead of being egocentric. They care about others’ reactions, which affects their anticipation to disapproval (Thompson, 2007; Thompson et al., 2006). Their ‘moral baggage grows into an inner moral sense,’ in which the abstract concepts of ‘care’ and ‘justice’ can be discerned (Schalkwijk, 2015, p. 27).
It is yet unclear to what extent attachment is associated with the different domains of conscience, because there is no integrated knowledge on the association between attachment and conscience. Notably, the meta-analysis by Hoeve et al. (2012) demonstrated that
attachment may be a morally relevant concept given its association with delinquency. However, it is still unknown whether the association between attachment and delinquency is mediated by moral development or conscience.
1.1.Attachment
The fact that conscience development begins during the child's early years emphasizes the importance of secure attachment relationships (Laible & Thompson, 2000). In child development theories, attachment refers to the extent to which the child approaches the caregiver in the event of stress, tension, anxiety, and illness, and the extent to which the caregiver is able to provide the child with a safe and familiar basis and haven that regulates stress and negative emotions (Bowlby, 1988; Hoeve et al., in press). In a secure attachment relationship, when the child and caregiver are well attuned to each other, the child on the one hand seeks proximity in times of stress and negative emotions, and on the other hand explores the world, encouraged by the parent to do so (Bowlby, 1988; Van IJzendoorn, 1994). A secure attachment relationship is crucial for the development of a child. When caregivers are
sensitive and responsive to the child’s attachment needs, and approach the child as a person who has his own thoughts and feelings, then the child will have a high chance of becoming securely attached to his or her caregiver(s) (Zeegers, Colonnesi, Stams, & Meins, 2017).
Based on parent-child experiences, the child builds an internal working model of attachment, which includes the ideas and thoughts that the child has about himself, the other, and the world (Bowlby 1973), reflecting the emotional quality of the relationship he/she has experienced, be it loving and warm, or mechanical and distant. Most children develop a
secure internal working model of attachment, characterized by children expressing their emotions to their primary caregiver, who provides comfort and acts as a secure base and safe haven for the child. However, when the attachment relationship lacks reciprocity, whereby the child experiences the caregiver as unavailable and insensitive, the child develops an insecure attachment style (Zeegers et al., 2017).
There are three insecure patterns of attachment types that can be distinguished: avoidant attachment, ambivalent/preoccupied attachment, and disorganized attachment (Ainsworth et al., 1978; Main & Salomon, 1990). An avoidant attachment style is
characterized by caring less about the self in relationships with others. Avoidantly attached children express their emotions least in the presence of their primary caregiver, because they have experienced that these emotions are ignored or rejected by the caregiver. Ambivalently attached children maximize their negative feelings in order to attract the attention of their inconsistent parents. Disorganized attached children have not developed a consistent strategy to regulate proximity and distance to their caregiver(s) in stressful situations, and therefore often exhibit unpredictable behavior. The behavior of these children can also be contradictory: for instance, when the parent leaves the child, the child can be very upset, but when the parent returns, the child can react indifferently. This attachment pattern can only be categorized when there is a history of evident trauma, and can be considered one of the most worrying forms attachment insecurity (Main & Salomon, 1990). These children often experienced abuse, neglect or abuse by the primary caregiver (Main & Salomon, 1990).
Traumatic experiences may lead to the development of a Reactive Attachment
Disorder (RAD; Carlson, et al., 1995; Lyons-Ruth & Jacobvitz 1999). A RAD can be defined as a disturbance in a child's relationships and social capabilities, which has arisen due to abuse and neglect by the primary caregivers (DSM-5; American Psychiatric Association [APA], 2013; Richters & Volkmar, 1996). It can be considered a disorder in the bonding of the child
with the caregiver (Richters & Volkmar, 1996). A child with a RAD shows socially and emotionally withdrawn behavior, which reduces learning opportunities, and limits the child's love and care from others (Prior & Glaser, 2006; Richters & Volkmar, 1996; Zeanah & Gleason, 2015). Furthermore, children with a RAD do not seek support and comfort when they are in distress or experience anxiety, and they hardly respond to the comfort that is offered to them (Zeanah & Gleason, 2015). In the case of a RAD, the child develops a
negative internal working model of attachment, based on adverse childhood experiences, and expects adults to respond in a negative way (Schuengel et al., 2003).
1.2.Moral emotions
Moral emotions are defined as self-evaluative, self-directed, self-conscious, or other-oriented emotions in response to morally significant situations that conform to or violate a moral standard (Levebvre & Krettenauer, 2019; Malti & Dys, 2015; Weiner, 2006). Moral emotions either assume the evaluation of one’s own actions and the effects these have on others, or the appreciation of others’ emotions (Malti & Dys, 2015). Moral emotions are closely tied to the self, and have a strong influence on one’s moral decisions. Moral emotions regulate social behaviors and evaluate one’s own as well as another’s actions as morally commendable or blamable or as positive or negative (Haidt, 2003, 2007; Kristjánsson, 2010; Tangney et al., 2007; Weiner, 2006).
Several researchers labelled empathy, shame and guilt as moral emotions (Haidt, 2003; Howe, 2013; Weiner, 2006). However, other researchers consider empathy not as an emotion, but rather as an ability, which leads to an emotion, to accurately perceive or experience the emotion of another person as separate from one’s own (e.g. Allen, Fonagy & Bateman, 2008; Aragno, 2008; Bolognini, 2004; Schalkwijk, 2015; van IJzendoorn, 1997). Despite the inconsistency in the definition of moral emotions, we consider empathy to be a moral
emotion, because it involves the ability to identify with the feeling of another individual(i.e., cognitive component) and vicariously share these emotions (affective component) (Howe, 2013).
1.2.1. Empathy
Empathy is the ability to perceive or show concern for another person’s thoughts or feelings (Jenkins, Demaray, Fredrick, & Summers, 2016; Schalkwijk, 2015). It is conceptualized as a fundamental factor of social competence and antecedent of moral reasoning and moral behavior (Davis, 1996; Eisenberg & Miller, 1987; Mehrabian, Young, & Sato, 1988; van IJzendoorn, 1997). Three aspects play a role in empathic ability: emotion, cognition and body (Schalkwijk, 2015). When a person wants to empathize with or understand another person’s feelings and thoughts, he has to use his own feeling, cognition and bodily reaction
(Schalkwijk, 2015). Empathy is also called a fundamental process of self-psychology, and thus an important aspect in the functioning of conscience (Schalkwijk, 2018). Self-evaluation can occur interpsychically in real interactions with another person or intrapsychically in interaction with an imaginary self-object (Schalkwijk, 2018).
Furthermore, empathy can be differentiated into cognitive and affective empathy (Hoffman, 2000; Meissner, 2009). Gladstein (1983) and Smith (2006) describe affective empathy as an automated, occasionally unconscious emotion resonating with another’s emotion, showing concern for another’s suffering. Cognitive empathy is described as a conscious, yet intellectual understanding of another’s emotion, which comes from taking the perspective of the other person (Gladstein, 1983 & Smith 2006). In contrast to the way conscience functions, the empathic ability can also be consciously switched on and off or function in between (Schalkwijk, 2018). The ability to resonate to and understand another person’s emotion is fundamental in forming and preserving social relationships, and starts early in life, especially within a family setting (Anderson & Keltner, 2002; Batson, 1991;
Preston & de Waal, 2002; Twemlow, Fonagy, & Sacco, 2005). Empathic ability is not only fundamental in interaction with others, but it is also important for the emergence of feelings of shame and guilt about fantasized interactions with imaginary others (Schalkwijk, 2015).
1.2.2. Shame
Shame is a negative moral emotion that relates to the negative self-evaluation of one’s whole being (e.g. “I am a dumb person;” Lewis, 1997; Schalkwijk, 2015; Van Kleef, Van Doorn, Heerdink, & Koning, 2011). Shame and guilt both act as intermediaries of the conscience, and are strongly associated with empathy, moral behavior and moral identity (Eisenberg, 2000; Hoffman, 2000; Tracy, Robins, & Tangney, 2007). The self-evaluated state combined with the understanding of moral rules and standards differentiate guilt and shame (self-conscious emotions) from empathy (Lewis, 2000).
Shame is the most powerful self-conscious emotion, as it is impossible to anticipate and quickly overwhelms (Schalkwijk, 2015). Shame gives one a feeling of losing control. If the person deals adaptively with shame, it becomes a bypassing emotion and disappears (Tracy, Robins, & Tangney, 2007). If shame, however, is dealt with maladaptively, chronic feelings of inferiority and failure may develop, resulting in more shame (Pines, 2008). Along with shame, people often experience feelings of anger when they feel devalued (Rodriguez Mosquera, Fischer, Manstead, & Zaalberg, 2008; Rodriguez Mosquera, Manstead, & Fischer, 2002). Some cope with the overwhelming character of shame and the experience of losing control by attributing the cause of this self-conscious emotion and anger to others (Elison, Lennon & Pulos, 2006; Nathanson 1992). Just as in the case of Joshua, who turned the resulting anger outside by tearing apart the drawing, to cover up the feelings of shame and restore his sense of ‘self’ (Schalkwijk et al., 2016). Joshua felt ashamed and devalued by his caregiver, and acted out his anger by tearing apart the drawing he had made for the caregiver.
Others cope with shame by blaming themselves and turning away from the other person, turning the anger against themselves (Schalkwijk, 2015). Even though shame originates from evaluation of one’s self, Joshua is ashamed because he is seen by the other (Schalkwijk, 2015). The feeling of losing control goes along with the feeling of being caught, which in turn causes him to attribute the feeling of shame to the caregiver: ‘the caregiver caused me to be ashamed’ (Schalkwijk, 2015).
1.2.3 Guilt
Guilt is also described as a negative moral emotion that relates to the negative evaluation of behavior (e.g., “I did not try hard enough;” Lewis, 1971; Van Kleef, Van Doorn, Heerdink, & Koning, 2011). Guilt pertains to one’s actions and the realization that actions, considered meaningful, influences another person (Lewis, 1971; Schalkwijk, 2015). Guilt, often
experienced as a feeling of distress, remorse, regret, sadness, unquietness and at times fear, is more reflective and less overwhelming compared to shame (Schalkwijk, 2015). Just like shame, guilt needs to be regulated (Tracy, Robins, & Tangney, 2007). Guilt is connected with empathic abilities, as it is about a person’s ability to empathize with another (Schalkwijk, 2015). Shame and guilt can be differentiated. The self-awareness in shame stems from a negative evaluation of the self by comparing who you are with who you think you or want to be, while guilt originates from the evaluation of one’s actions. Both shame and guilt can occur relatively to an internalized, absent other, who is not aware of the person’s actions
(Schalkwijk, 2015).
1.3.The association between attachment and moral emotions
Researchers have examined the link between attachment (in)security1 and the moral emotions of empathy, shame, and guilt in childhood, adolescents and young adults (e.g., Joireman, Needham, & Cummings, 2002; Kestenbaum et al., 1989; Lopez et al., 1997; Muris
et al., 2014; van der Mark et al., 2002; Sroufe, 2005). Literature shows that secure attachment is related to higher levels of empathy and lower levels of (maladaptive) shame and guilt compared to an insecure attachment pattern (Muris, et al., 2014; van der Mark et al., 2002). In contrast, Akbağ and Erden İmamoğlu (2010) found that shame was predicted by a secure and avoidant attachment styles. Furthermore, other researchers found that children with an
avoidant attachment style displayed in between levels for shame (neither different from secure nor ambivalent attachment) and showed lower levels of (cognitive) empathy than
ambivalently attached children (Muris et al., 2014; Ştefan, 2019). In addition, secure and avoidant attached young adults showed lower levels of shame than their preoccupied and fearful counterparts (Gross & Hansen, 2000; Passanisi et al., 2015). However, Akbağ and Erden İmamoğlu (2010) found no relation between preoccupied attachment and shame compared to the other attachment patterns. In addition, Muris et al. (2014) found that alienation was related to higher levels of maladaptive forms of shame and guilt. Besides, Akbağ and Erden İmamoğlu (2006) found that more avoidant attachment was related to less guilt.
Attachment is characterized by two underlying dimensions: a positive or negative model of the self and others (Bartholomew, 1990; Bowlby, 1973). Securely attached individuals are more likely to show empathy, because a positive internal working model of attachment is associated with increased mentalizing abilities (Zeegers, Colonnessi, Stams, & Meins, 2019). Secure individuals are also assumed to experience relatively low levels of shame and higher levels of guilt because of their positive internal working model of the self and others (Bartholomew & Horowitz, 1991). Avoidantly attached individuals fear and avoid emotional closeness as a self-protection mechanism, with is incompatible with empathic concern and feelings of guilt and shame (Akbağ & Erden İmamoğlu, 2010; Gillath, Shaver, &
1Different attachment styles and patterns are used in literature during infancy, childhood and adolescence. Even though the constructs are not all the same, yet the patterns are most likely the same across the life course. In this study we chose to use the original attachment categories secure, avoidant, and ambivalent from the Strange Situation (Ainsworth et al., 1978), to cover the different patterns of attachment. Secure (Type B): closeness, communication, trust; Avoidant or dismissive/avoidant (Type A): dismissive, avoidant, fearful, alienation; Ambivalent or ambivalent/preoccupied (Type C): ambivalent, preoccupied, anxiety, dependent, resistant
Mikulincer, 2005; Mikulincer & Shaver, 2005; Richman, DeWall & Wolff, 2015).
Ambivalent attachment represent a negative self-model, which is related to shame proneness, considering the intrinsic doubt about others’ availability and responsiveness to attachment needs (Bartholomew & Horowitz, 1991; Colonnesi et al., 2011; Cook, 1992; Spruit et al., 2020). Shame also seems to have a self-destructive effect, and therefore mainly seems to be an internalized reaction to perceived (or inferred) self-devaluation in relation to others
(Colenessi et al., 2011; Tangney & Dearing, 2002). Notably, the relation between attachment and self-conscious emotions is complex, particularly with regard to shame, because it seems to be the way people cope with shame that determines whether shame is an adaptive reaction or not that can possibly protect against delinquency (Muris et al., 2014; Schalkwijk, Dekker, et al., 2016; Schalkwijk, Stams et al., 2016).
1.4 The present meta-analysis
The main aim of the current meta-analysis was to examine the relation between attachment and moral emotions. A second aim was to examine moderators that might influence the relation between attachment and moral emotions, accounting for attachment characteristics, attachment figure, attachment assessment, moral emotions assessment, and participant and study characteristics. We hypothesized that (1) secure attachment would be related to more empathy and less maladaptive forms of shame and guilt compared to insecure attachment, and (2) avoidant/dismissive attachment would be related to lower levels of empathy, shame and guilt compared to secure attachment.
2. Methods
2.1. Selection of studies
We aimed to include all studies on the relation between attachment and moral development in the current meta-analysis. The included studies were found in three consecutive steps. The only inclusion criterion was that the study should describe an association between moral reasoning or moral judgment, empathy or shame and guilt (moral emotions), and attachment to parents, peers, partners or attachment based on the internal working model. Only sample types from early childhood to young adults were included. The first step was to do a broad advanced search through the databases Ovid, ERIC (Educational Resources Information Centre) and PsycINFO with the following search terms: attachment, empathic or empathy or shame or guilt, and moral reasoning or moral judgment. For the complete search string, see appendix.
These databases collected 485 studies that all focus on moral reasoning, moral
judgment, empathy and shame and guilt. All titles, abstracts, and full-text articles found were screened and excluded if they did not meet the inclusion criterion. The literature search included all studies that were found until April 2020. From this point on we focus only on empathy, shame, and guilt, because Koster (2020) focused on moral judgement. In the second step studies were located using the snowball method, that is inspection of the reference
sections of the articles, the narrative reviews and book chapters. Finally, 21 studies met the inclusion criterion that studies had to examine the association between attachment to parents, caregivers, partners, peers and empathy, shame, or guilt.
On closer inspection, one study had to be excluded, because study included attachment and moral maturity and empathy but didn’t include the correlation between attachment and empathy. We tried to request the correlations from the author, but did not receive the information (Holmqvist, 2008).
Eventually, 20 studies met the inclusion criteria that studies had to examine the association between attachment to parents, caregivers, peers, partners or attachment based on the internal working model and moral emotions. The following has been excluded from this meta-analysis: reviews, qualitative studies, studies with no statistical information,
inappropriate age, animal studies, parental and maternal attachment/empathy, disorders (e.g. autism, depression, borderline), internalizing problems, psychopathy, bullying/victimhood, instrument descriptive studies, biological/neurological studies and mindfulness.
Finally, we included 20 studies, based on 22 independent datasets, only published ones, between 1988 and 2019, which examined 142 associations between attachment and empathy or shame or guilt. Both longitudinal (k = 3) as well as cross-sectional (k = 19) studies were included, with all kinds of different sample types, such as residential youths, community samples, criminal offenders, students, adoptive or foster care children and
children living in Israelian Kibbutzim. In total, data of 7,457 participants were analyzed in the present study. Sample sizes ranged from 59 (Oppenheim, Sagi & Lamb, 1988) to 2834
(Acosta et al., 2019) participants, with an average of 310 participants per dataset. The age of the adolescents ranged from age 1 – 23 with a mean age of 18.75 years. There were some outliers in age in samples of higher education or university students. Table 1 presents the included studies and their main characteristics. Included studies in the meta-analysis are marked with * in the references.
Table 1.Characteristics of included studies.
Author (Year) N IV Peer
review
IF Design Informant Continent Type attach
Sex Ethnic Min.
Ag e Type sample
Panfile and Laible (2012) 63 Empathy Yes 1.23 Cross Both NA B Mixed .19 3 Normal
Yu, Wang, and Liu (2012) 559 Empathy Yes 1.36 Cross Child Other B Mixed 1 20 Normal
Termini, Golden, Lyndon, and Sheaffer (2009) 73 EmpGuilt Yes 2.02 Cross Child NA RAD Mixed .44 10 Mixed
Oppenheim (1988) 59 Empathy Yes 2.93 Long Other Other C Mixed 1 5 Risk
Zarrella, Lonigro, Parrella, Caviglia, and Laghi (2016) 159 Empathy Yes 0.48 Cross Child Europe AC Mixed 1 10 Normal
Laible (2007) 117 Empathy Yes 1.97 Cross Child NA B Mixed .22 20 Normal
Muris et al. (2014a) 688 GuiltSham Yes 1.78 Cross Child Europe AC Mixed .10 10 Normal
Muris et al. (2014b) 102 GuiltSham Yes 1.78 Cross Child Europe AB Mixed .10 15 Risk
You and Kim (2016) 730 Empathy Yes 1.48 Cross Child NA B Mixed – 15 Normal
Acosta, Chinman, Ebener, Malone, Phillips and Wilks (2019) 2834 Empathy Yes 1.96 Cross Child NA B Mixed .14 11 Normal
Panfile and Laible (2013) 71 Empathy Yes 2.02 Cross Parents NA B Mixed .09 3 Normal
Diamond, Fagundes and Butterworth (2011) 114 Empathy Yes 2.07 Long Child NA AC Mixed .18 14 Normal
Van der Mark, Ijzendoorn and Bakermans-Kranenburg
(2002) 151 Empathy Yes 2.04 Long Parents Europe B Female – 1 Normal
Joireman, Needham and Cummings (2002a) 134 Empathy Yes 1,87 Cross Child NA BC Mixed .08 19 Normal
Joireman, Needham and Cummings (2002b) 261 Empathy Yes 1,87 Cross Child NA ABC Mixed .08 19 Normal
Lopez et al. (1997) 142 GuiltSham Yes 1,41 Cross Child NA AC Mixed .30 22 Normal
Ştefan and Avram (2019) 212 Empathy Yes 1,68 Cross Child Europe AC Mixed – 5 Normal
Passanisi et al. (2015) 215 Shame Yes 0,78 Cross Child Europe ABC Mixed – 22 Normal
Wei et al. (2005) 299 Shame Yes 3,34 Cross Child NA AC Mixed .19 20 Normal
Gross and Hansen (2000) 204 Shame Yes 1,97 Cross Child NA ABC Mixed .11 22 Normal
Wei (2011) 195 Empathy Yes 2,94 Cross Child NA AC Mixes .05 23 Normal
Abdullah et al. (2014) 75 GuiltSham Yes 0,78 Cross Child Other B Female – 16 Risk Note. N = number of participants; peer review = published in peer reviewed article yes/no; IF = impact factor of journal; design = cross-sectional or longitudinal; Informant = informant of attachment measure; Continent = location of study; IV = independent variable; Ethnic min. = proportion non-Caucasian; Cross = cross-sectional design; Long = longitudinal design; GuiltSham= Guilt & Shame; EmpGuilt = Empathy & Guilt; Child = data from child; Parents = data from parents; Both = data from parent and child; Other = data from other source; NA = North America; Other (continent) = Asia (China, Malaysia & Israel); A = Avoidant attachment; B = Secure attachment; C = Ambivalent, Resistant, Anxious attachment; RAD = Reactive attachment disorder; Age = mean age; Normal = non-risk sample; Risk = Risk sample; Mixed = Non-risk and risk sample.
2.2. Coding and moderators
Included studies were coded according to the guidelines of Lipsey and Wilson (2001).
Potential moderators of the association between attachment and moral emotions were grouped into attachment characteristics, attachment figure, attachment assessment, empathy, shame and guilt assessment, participant characteristics and study characteristics. Trust,
communication, and closeness were recoded into secure type B attachment; dismissive, avoidant, and alienation were recoded into avoidant type A attachment; ambivalent,
preoccupied, fearful, anxious, dependent, and resistant were recoded into ambivalent type C attachment. RAD and disorganized type D attachment were not recoded.
With whom the attachment relationship had been measured was categorized under attachment figure as follows: with parents, peer/partner, or unspecified. As attachment
assessment, questionnaires or observations (AQS, AQC, SAT, IPAA, Strange situation, IPPA) and interviews (DSM) were used, in a continuous (e.g., more or less insecure) or categorical
(e.g., secure, ambivalent, or avoidant) type of attachment measure. Self report ( The Basis Empathy Scale, case vignettes, Faste, IRQ, SCEMAS, IRI, SSIS-RS, discussion task), Objective measure (IAT, Procedure outlines, Simulations) and another type (not specified; My Child) instruments were used to measure empathy, shame or guilt. As participant characteristics, percentage male and age were coded. The sample types consist of general populations (RC) and at-risk populations, and as study characteristics publication year, impact factor, published (yes or no) and continent (Northern America (RC) and Europe) were coded. Moral emotions were coded as the independent variable, and attachment was coded as the dependent variable. Univariate moderator analyses were performed to investigate which moderators significantly influenced the association between attachment and empathy or self-conscious emotions. Of the final model, it was investigated whether the variance at level 2 and 3 significantly deviated from zero. Next, it was again established how much of the total variance was found at each level. Finally, it was calculated how much more of the proportion of variance was explained by adding moderators, using the formulas of Cheung (2014). All analyses were performed in R with the metafor and the glmulti package (Viechtbauer, 2010; 2020).
2.3.Calculation and analysis of effect sizes
A meta-analyze was conducted to assess the relation between attachment and moral emotions. For the effect size, correlation coefficients were calculated using formulas from Lipsey and Wilson (2001). All correlation coefficients were drawn from bivariate relations or from partial correlations that controlled for the covariance of moral emotions. To correct for dependency because of multiple effect sizes derived from the same studies, a three-level random effects model was used for the analyses, so there were three levels of variance (Field, 2001),
including the sampling variance for each effect sizes (level-one), the variance between effect sizes within studies (level-two), and the variance between studies (level-three) (Wibbelink &
Assink, 2015). We used likelihood-ratio-tests to compare the deviance scores of the full model and the models without variance parameters on level two or three to determine if the level-two and -three variances were significant, indicating heterogeneous effect sizes. In case the effect sizes are considered to be heterogeneous, we proceeded to moderator analysis, since the differences between the effect sizes may be explained by characteristics related to
attachment, empathy and self-conscious emotions, and samples. The current meta-analysis was conducted in R, using the metafor-package, employing the restricted maximum likelihood estimation and the Knapp and Hartung-method (Viechtbauer, 2010; Weisz et al., 2013;
Wibbelink & Assink, 2015).
2.4. Publication bias
In meta-analyses, the aim is to include all studies previously conducted that meet the inclusion criteria (Lipsey & Wilson, 2001). However, some studies may not have been published due to non-significant or unfavorable results, and therefore difficult to locate. This may result in the so-called “file drawer bias”, and can lead to stronger estimations than the true effect size (Rosenthal, 1995). First, we tested for funnel plot asymmetry by regressing the standard normal deviate (effect size divided by standard error) against the effect size's precision (inverse of the standard error) in SPSS (Egger, Davey Smith, Schneider, & Minder, 1997). After that, a trim-and-fill analysis was performed to investigate whether adding effect sizes to the right or the left side of the distribution influenced the mean estimate (e.g., under- estimation or overestimation) of the overall effect size (Duval & Tweedie, 2000a, 2000b). These analyses were performed with a two- level random-effects model.
3. Results
The meta-analysis included 20 studies (k), with 22 independent samples, 142 effect sizes, and a total of 7,457 participants. The results of the multilevel meta-analysis on the relation
between attachment and moral emotions in children, adolescents, and young adults are presented in Table 2 and 3. The overall association between attachment and moral emotions can be found in Table 2.
3.1. Overall effect size
Table 2 shows that attachment proved to be significantly associated with moral emotions, showing a small effect size of r = .153 (t = 3.257, p < .001). There was a significant within study variance of 96%. Therefore, we conducted moderator analyses to explain within study differences in effect sizes.
Table 2. Overall Relation between Attachment and Moral Emotions
Outcome k #ES Mean
r
95% CI p σ2level 2 σ2level 3 % Var.
Level 1 % Var. Level 2 % Var. Level 3 Youth- outcomes 22 142 .153 .060; .242 .001 0.254*** 0.005 1.861 96.343 1.796 Note. Youth outcomes = academic and vocational, social-emotional, physical health,
psychosocial problems; k = number of independent samples; #ES = number of effect sizes; mean r = mean effect size (r); CI = confidence interval; σ2
level 2 = variance between effect
sizes extracted from the same study; σ2
level 3 = variance between studies; % Var = percentage
of variance distributed.
*p < .05 **p < .01 ***p < .001
3.2. Publication bias and Moderator analysis
The results of the trim-and-fill-analysis suggested that bias was present in the data set, because of an asymmetric distribution of effect sizes. From the funnel plot in Figure 1, it can be derived that effect sizes were missing on the right sight of the funnel, and consequently, 51
effect sizes (from 7 studies) were added to the dataset, so that a “corrected” overall effect could be estimated of r = .344, t = 4.600, p < .001.
Table 3 presents the results of the moderator analyses on the relation between attachment and empathy, guilt or shame. It can be derived from Table 3 that only one moderator was significant, namely, type of attachment measure. Studies that used a categorical type of attachment measure yielded larger effect sizes (r = .241), compared to continuous type of attachment measure (r = .063). A trend significant effect was found, indicating that in general population samples attachment was (moderately) associated with moral emotions (r = .196), whereas in at risk samples no significant association was found.
Fig. 1. Funnel plot with Fisher's z transformed correlations. On the y-axis are the standard
errors of the studies, with smaller standard errors representing larger sample sizes. On the x-axis are the associations between attachment and moral emotions.
Table 3. Moderating Variables of the Relation between Attachment and Moral Emotions
Moderator variable k #ES B0/
Fisch. z r t0 B1 t1 F(df1, df2) Attachment characteristics Attachment type F(3, 137) = 0.261 RAD 1 8 .106 .106 0.527 Avoidant 11 34 .209 .206 2.269* .104 0.470 Ambivalent/preoccupied 12 41 .167 .165 1.981* .061 0.282 Secure 13 58 .114 .114 1.557 .008 0.038 Attachment figure F(1, 140) = 1.370 Parents 14 71 .200 .197 3.201** Peers or partner 13 71 .181 .179 1.196 -.103 -1.170 Attachment assessment Type of instrument F(2, 139) = 0.102 Questionnaire 16 118 .149 .148 2.764** Interview 2 14 .219 .216 1.458 .070 0.441 Observation 4 10 .140 .139 0.819 -.009 -0.049
Type of attachment measure F(1, 140) = 4.624*
Continuous 10 78 .063 .063 1.088 Categorical. 13 64 .248 .241 3.877*** .185 2.150* Moral Emotions Moral concept F(2, 139) = 0.066 Empathy 15 70 .155 .154 2.320* Shame 7 35 .135 .134 1.440 -.020 -0.174 Guilt 5 37 .180 .178 1.899+ .025 0.214 Type of instrument F(2, 139) = 0.058 Questionnaire self-report 18 128 .157 .156 3.067** Questionnaire other-report 2 5 .202 .199 0.845 .045 0.185 Objective measure 3 9 .105 .105 0.583 -.052 -0.276 Participant characteristics % male sample 22 142 .160 .159 3.261** .324 1.182 F(1, 140) = 1.397 Age 22 142 .154 .153 3.262** -.002 -1.149 F(1, 140) = 1.319 Sample type F(1, 140) = 3.143+ General population 19 96 .199 .196 3.792*** At risk population 3 46 .033 .033 0.427 -.166 -1.773+ Study characteristics Publication year 22 142 .155 .154 2.249** .006 0.820 F(1, 140) = 0.673 Impact factor 22 142 .155 .154 3.229** .008 0.085 F(1, 140) = 0.007 Study design F(1, 140) = 0.000 Cross-sectional 19 127 .155 .154 3.020** Longitudinal 3 15 .158 .157 1.106 .003 0.018 Continent F(2, 139) = 0.363 Northern America 13 68 .124 .179 1.780+ Europe 6 59 .212 .204 2.603* .088 0.823 Other 3 15 .128 .283 0.876 .004 0.022
Note. IV and DV Characteristics = independent variable (IV) and/or dependent variable (DV); k = number of independent studies; #ES = number of effect sizes; B0/ mean r = intercept/ mean effect size
(r); t0 =difference in mean r with zero; B1 =estimated regression coefficient; t1 = difference in mean r
with reference category; F(df1, df2) = omnibus test.
4. Discussion
The purpose of the present meta-analysis was to investigate the association between attachment and moral emotions in order to obtain a better understanding of the relation between attachment and delinquency. In addition, we tested whether moderators influenced the strength of this association. An overall small but significant association was found between attachment and moral emotions. A moderate and significant association was found between attachment and moral emotions for studies using categorical attachment measures, and a non-significant association for studies using continuous attachment measures. The association between attachment and moral emotions was not moderated by attachment characteristics, attachment figure, type of attachment instrument, type of moral emotion (i.e., shame, guilt, or empathy), participant characteristics, sample type, and study characteristics. However, a trend was found, which indicated that sample type affected the association between attachment and moral emotions in that general population samples yielded a significant and moderate association, whereas samples of at-risk youth did not produce a significant effect. Finally, the association between attachment and moral emotions may be a substantial underestimation of the true effect, and therefore the magnitude of the overall effect size should be interpreted with some caution.
The finding that the relation between attachment moral emotions is modest is not in line with results from the meta-analysis of Koster (2020), who found a very small association between attachment and moral judgement. The weak association between attachment and moral judgment may be explained by the fact that attachment is a cognitive-affective working model of relationships with important others, while moral judgment mostly pertains to
cognition, although it might be influenced by moral emotions, such as empathy, shame, and guilt, which motivate moral behavior (Tangney, Stuewig & Mashek, 2007; Zahn-Waxler, et al., 1992). This dual process of moral emotions and moral judgement is also supported by the
fMRI study of Green et al. (2001), which shows that areas in the brain associated with emotion are more activated when one is reflecting on personal moral dilemmas. Moral
judgement has an effect on behavior when it connects with moral emotions (Chapman, 2009). As Pizarro (2000, pp. 370-371) states “The Capacity to experience empathy and the ability to regulate it efficiently are necessary in order to be a moral individual.” Moral emotions and moral judgement together cause (moral) behavior. Another explanation for differences in the strength of the association between attachment and moral cognition (i.e., judgment) and moral emotions is that the meta-analysis by Koster (2020) predominantly included samples of late adolescents and young adults instead of children and early and middle adolescents, given that the role of attachment in relationships with other(s) is thought to change across the life course (Howe, 2011).
The moderator analyses showed a significant effect for type of attachment measure (categorical vs. continuous). Studies using categorical measures yielded a significant effect size, whereas no significant effect was found for studies using continuous measures. This is a somewhat surprising effect, as literature shows that continuous attachment measures in general yield larger effect sizes than categorical measures, and are better suited for conceptualization and measuring individual differences in attachment (e.g., IJzerdoorn & Kranenburg, 2014; Joireman et al., 2002; Fraley, et al., 2015; Spruit et al., 2020). Although most researchers recommend the use of continuous attachment measures, Both and Best (2017) argue that also categorical measures have advantages, because of high face validity given that many researchers conceptualize individuals as being secure, fearful, preoccupied, or dismissing, and for instance scoring high on both avoidant and preoccupied/ambivalent seems counterintuitive. Moreover, the golden standard of assessing attachment is still the Strange Situation procedure and the Adult Attachment Interview, which are categorical measures that according to many scholars in the field of attachment research have the highest
degree of validity (Solomon & George, 2008; Van IJzendoorn, 1995). Based on these
inconsistencies, more research is needed to examine the impact of categorical and continuous attachment measures and other assessment characteristics on results of attachment studies (see Madigan, Brumariu, Villani, Atkinson, & Lyons-Ruth, 2016). Notably, Spruit et al. (2020) highlights the importance of using multiple type of measures to assess attachment.
The finding that the association may be a substantial underestimation of the true effect can be explained by the fact that larger effect sizes may be expected if moral functioning is considered as a whole, without tearing apart the different constituents of the conscience, that is, moral emotions and moral cognition. The conscience, a psychological function, results from an interplay between empathy, self-conscious emotions, such as shame and guilt, and moral reasoning, and monitors the balance of self-esteem and identity (Schalkwijk, 2018). The development of empathy, shame, guilt, and moral reasoning affects internalized standards of behavior (Thompson, 2007; Thompson et al., 2006). Therefore these constructs taken together might yield a larger effect size.
Findings on the association between attachment and moral development or conscience (i.e., moral emotions and moral cognition) may increase our understanding of the established link between attachment and delinquency (Hoeve et al., 2012). In the meta-analysis of Stams et al. (2006) on moral judgement and juvenile delinquency a stronger medium-to-large association was found between moral cognition and delinquency. The small association between attachment and moral judgment in the meta-analysis of Koster (2020) renders it unlikely that the relation between attachment and delinquency would be mediated by moral cognition. The meta-analyses of Spruit et al. (2016) on the relation between self-conscious emotions (guilt and shame) and delinquency and Van Langen et al. (2014) on the relation between empathy and juvenile delinquency showed small to moderate associations for the relation between moral emotions and delinquency. The modest association between
attachment and moral emotions (the present meta-analysis), which may even be a substantial underestimation of the true association, suggests that the relation between attachment and delinquency may rather be mediated by moral emotions than moral cognition. Figure 2 summarizes the mediation model of attachment, moral development, and delinquency.
Fig. 2. Mediation model of attachment and delinquency/criminal recidivism
4.1. Limitations, strengths and recommendations
There are some limitations to this meta-analysis that need to be mentioned. The first limitation involves the conceptualization of empathy. There is a lack of a clear and agreed upon
definition of this construct. Several researchers claim empathy to be a moral emotion, whereas others do not consider it to be a moral emotion, but rather an ability to perceive another’s feelings (e.g. Aragno, 2008; Bolognini, 2004; Haidt, 2003; Howe, 2013; Weiner, 2006; Schalkwijk, 2015). Both conceptualizations of empathy are censored and fail to capture the full range of the construct. Simultaneously, empathy is a complex construct (Coplan, 2011). In our meta-analysis some studies conceptualized empathy in terms of cognitive and affective empathy, perspective taking (cognitive), behavioral empathic perspective taking (cognitive),
and personal distress (e.g., Diamond, Fagundas, & Butterworth, 2011; Joireman et al., 2002), whereas other studies conceptualized empathy without such distinctions (Panfile & Liable, 2012; You & Kim, 2016). This might have influenced our results. For example, we also coded cognitive and affective empathy (Schalkwijk, Cima et al., in press). If cognitive empathy would not be considered as a moral emotion, it should have been included in the meta-analysis of Koster (2020) on the association between attachment and moral cognition, which would have somewhat increased the overall effect size.
A second limitation is the conceptualization of shame. Several researcher made a distinction between different forms of shame that we were unable to take into account in the current meta-analysis (Cohen et al., 2011). Positive and negative effects of shame are used interchangeably. Only one study in our meta-analysis made a distinction between ambiguous (negative) and un-ambiguous (positive) shame (Muris et al., 2014). Shame is a complex construct, as the adaptive or maladaptive outcome not only depends on its conceptualization, but also on how one copes with it. To explain this further, individuals who avoid emotional closeness (avoidant attachment ) will, as a self-protection mechanism, feel less shame, and in order to protect themselves against negative emotions in relationships with others, they will most likely use avoidant coping strategies (to avoid a situation or feeling), which is
predominantly related to shame and internalizing problems (Yi & Baumgartner, 2011; Mikulincer, Florian, & Weller, 1993). Simultaneously, internalizing problems, including internalized shame, can serve as a protective factor against delinquency (Schalkwijk, Dekker, Peen & Stams, 2016; Wibbelink, Hoeve, Stams & Oort, 2017). Therefore, the way of coping (a form of self-regulation to deal with negative emotions or experiences) with shame seems crucial, more so than coping with empathy or guilt, as it is associated with worse
psychological adjustment and maladaptive behavior (Compas et al., 2001; Ebata & Moos, 1991; Gross, 1998; Tangney, Wagner, and Gramzow, 1992).
Third, the different organized attachment patterns from infancy to (young) adulthood have been clustered into three categories: secure, avoidant, and ambivalent attachment, besides disorganized attachment. Fraley (2002) found that attachment security is moderately stable from infancy to 19 years of life, and can therefore change across the course of life. Change in attachment patterns later in life can occur through new attachment relationships, in which one can co-construct new attachment representations, which may lead to a complete change in one’s original model of attachment (Bowlby, 1969). Research on adult attachment has mainly been based on individual differences in infant and adult attachment patterns and representations (e.g., Hazan & Shaver, 1987; Main, Kaplan, & Cassidy, 1985; Mikulincer & Shaver, 2007). For instance, ambivalent attachment during childhood corresponds with preoccupied attachment during adulthood. It is not the same construct, but rather a same type of attachment pattern in a different type of relationship (child-parent vs adult-adult) (Main & Goldwyn, 1985; Shaver et al., 1988; Weiss, 1991). Several researchers found evidence that the specific attachment representations of a current adult relationship develop over time and are integrated into a generalized representation, which develops earlier within the child-parent relationship (Crowell, Fyffe, Gao, Pan, Treboux, & Waters, 2002; Crowell, Treboux, & Waters, 2002; Treboux, Crowell, & Waters, 2004; Zeifman & Hazan, 1997). However, there are different ideas about the origin of adult attachment, and there also seems to be
disagreement about attachment structure in adulthood, which causes confusion about the conceptualization of (adult) attachment (Crowell, Fraley, & Shaver, 2008). The three different categories in the current study do not fully capture the different attachment patterns,
attachment representations and attachment working models from infancy to (young)
adulthood. Probably this meta-analysis would have yielded a larger effect size if the problem in the conceptualization of attachment would not have been present.
Fourth, because of the limited amount of at risk samples and the limited variation among studies, there might be restriction of range problems. Last, moral emotions and moral judgement have been studied separately. Moral emotions and moral judgement may work together to bring forth moral behavior (Tangney, Stuewig, & Mashek, 2007; Zahn-Waxler, et al., 1992).
Despite these limitations, this meta-analysis has several strengths. This meta-analysis is the first to investigate the relation between attachment and moral emotions, and
simultaneously increases our understanding of the relation between attachment and moral functioning, and between moral functioning and delinquency. Secondly, the use of multi-level analysis made it possible to include all effect sizes on the relation between attachment and moral emotions with their distinctive possibly moderating characteristics, simultaneously correcting for statistical dependency.
The findings of the current study offer important implications for future research and clinical practice. Moral judgement and moral emotions should both be integrated in future research on attachment, moral functioning, and delinquency/criminal recidivism. More research is needed to test the mediation model as presented in this discussion. With regard to shame, more research is needed on how attachment influences the way people cope with shame. A more thorough and comprehensive understanding of shame and the coping thereof is needed to prevent delinquency and criminal recidivism in children and young adults, and may help clinicians in developing more precise measures of assessment and interventions, perhaps resulting in more effective treatment.
5. Conclusion
This meta-analysis provides a valuable exploration of the relation between attachment and moral emotions, and provides more insight in the relation between attachment and
delinquency, and the possible mediating role of moral functioning. This study has demonstrated that there is an association between attachment and moral emotions. It is therefore recommendable to treat children and young adult as moral individuals, without distinguishing moral emotions from moral judgment, and stimulate (positive) moral emotions, which in turn can influence moral judgement and motivate moral behavior (Tangney, Stuewig & Mashek, 2007; Zahn-Waxler, et al., 1992). Special awareness is needed with regard to shame, because of its often maladaptive outcomes and complexity. Interventions that target the improvement of coping styles in dealing with shame are recommendable. Furthermore, attachment should be assessed with multiple indicators. For children or adolescents with complex problems, in particular those with adverse childhood experiences, it is recommended to foster secure attachment relationships with caregivers through attachment-based
interventions or in foster care and family group homes (Spruit et al., in press; Spruit et al., 2020). Finally, the new mediation model for attachment and delinquency/criminal recidivism can expand existing clinical knowledge, and may serve as the subject of future research.
References
*Abdullah, S., Ghani, S., Akil, S., & Faudzi, N. (2014). Relationship of Parent and Peer Attachment with Coping Strategy among Teenagers Pregnancy. Procedia - Social and Behavioral Sciences, 114(C), 334–338. https://doi.org/10.1016/j.sbspro.2013.12.707 *Acosta, J., Chinman, M., Ebener, P., Malone, P., Phillips, A., & Wilks, A. (2019).
Understanding the relationship between perceived school climate and bullying: A mediator analysis. Journal of School Violence, 18(2), 200–215. https://doi.org/10.1080 /15388220.2018.1453820
Ainsworth, M. D., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-III-R). American Psychiatric Pub.
Akbag, M., & Imamoglu, S. (2010). The Prediction of Gender and Attachment Styles on Shame, Guilt, and Loneliness. Educational Sciences : Theory & Practice, 10(2), 669 –682.
Allen, J. G., Fonagy, P., and Bateman, A. W. (2008). Mentalizing in Clinical Practice. Washington, DC: American Psychiatric Publishing.
American Psychiatric Association (APA) (2013). Diagnostic and statistical manual of mental disorders. (5th ed) (DSM-5). Washington DC: Author
Anderson, C., & Keltner, D. (2002). The role of empathy in the formation and maintenance of social bonds. Behavioral and Brain Sciences, 25(1), 21–22. https://doi.org/10.1017 /S0140525X02230010
Aragno, A. (2008). The language of empathy: an analysis of its constitution, development, and role in psychoanalytic listening. Journal of the American Psychoanalytic Association 56(3), 713 – 740 https://doi.org/10.1177/0003065108322097
Bar-Haim, Y., Dan, O., Eshel, Y., & Sagi-Schwartz, A. (2007). Predicting children’s anxiety from early attachment relationships. Journal of Anxiety Disorders, 21(8), 1061–1068.
https://doi.org/10.1016/j.janxdis.2006.10.013
Bartholomew, K. & Horowitz, L.M. (1991). Attachment styles among young adults: A test of a four- category model. Journal of Personality and Social Psychology, 61(2), 226-244.
https://doi.org/10.1037/0022-3514.61.2.226
Bartholomew, K. (1990). Avoidance of Intimacy: An Attachment Perspective. Journal of Social and Personal Relations, 7(2), 147–178. https://doi.org/10.1177
/0265407590072001
Batson, C. D. (1991). The altruism question: Toward a social-psychological answer. Hillsdale, NJ: Erlbaum.
Bögels, S., & Brechman-Toussaint, M. (2006). Family issues in child anxiety: Attachment, family functioning, parental rearing and beliefs. Clinical Psychology Review, 26(7), 834–856. https://doi.org/10.1016/j.cpr.2005.08.001
Bolognini, S. (2004). Misunderstandings on empathy. Bull. Br. Psychoanal. Soc. 40, 1–13.
Bowlby, J. (1969). Attachment and loss: Vol. 1 attachment. New York: Basic Books. Bowlby, J. (1973). Attachment and loss: Volume II: Separation, anxiety and anger. The
International Psycho-Analytical Library (pp. 1-429). The Hogarth Press and the Institute of Psycho-Analysis.
Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. Basic Books.
Carlson, V., Cicchetti, D., Barnett, D., & Braunwald, K. (1995). Finding order in
disorganization: Lessons from research on maltreated infants’ attachments to their caregivers. In D. Cicchetti & V. Carlson (Eds.), Child maltreatment: theory and
research on the causes and consequences of child abuse and neglect (pp. 135–157). Cambridge University Press
Cheung, M. W. L. (2014). Modeling dependent effect sizes with three-level meta-analyses: A structural equation modeling approach. Psychological Methods, 19, 211–229. https://doi.org/10.1037/a0032968.
Colonnesi, C., Draijer, E. M., Stams, G. J. J. M., Van der Bruggen, C. O., Bögels, S. M., & Noom, M. J. (2011). The relation between attachment and child anxiety: a meta-analytic review. Journal of Clinical Child and Adolescent Psychology, 40, 630-645.
https://doi.org/10.1080/15374416.2011.581623
Compas, B. E., Connor-Smith, J. K., Saltzman, H., Thomsen, A. H., & Wadsworth, M. E. (2001). Coping with stress during childhood and adolescence: Problems,
progress, and potential in theory and research. Psychological Bulletin, 127, 87–127. https://doi.org/10.1037/0033-2909.127.1.87
Cook, D.R. (1992). Shame, attachment, and developmental psychopathology: A review with research and clinical implications. Unpublished manuscript, University of Wisconsin-Stout.
Crowell, J., Fraley, R. C., & Shaver, P. R. (2008). Measures of individual differences in adolescent and adult attachment. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd Edition)(pp. 599-634). New York: Guilford Press.
Crowell, J. A., Treboux, D., & Waters, E. (2002). Stability of attachment representations: The transition to marriage. Developmental Psychology, 38, 467-479.
Crowell, J. A., Treboux, D., Gao, Y., Fyffe, C., Pan, H., & Waters, E. (2002). Assessing secure base behavior in adulthood: Development of a measure, links to adult
attachment representations, and relations to couples’ communication and reports of relationships, Developmental Psychology, 38, 679-93. https://doi.org/10.1037/0012 -1649.38.5.679
Davis, M. H. (1996). Empathy: A Social Psychological Approach. Boulder, CO: Westview Press.
*Diamond, L. M., Fagundes, C. P., & Butterworth, M. R. (2012). Attachment style, vagal tone, and empathy during mother–adolescent interactions. Journal of
Research on Adolescence, 22, 165–184. https://10.1111/j.1532-7795.2011.00762.x Duval, S., & Tweedie, R. (2000a). Trim and fill: A simple funnel-plot–based method of
testing and adjusting for publication bias in meta-analysis. Biometrics, 56(2), 455–463. https://doi.org/10.1111/j.0006-341X.200.00455.x.
Duval, S., & Tweedie, R. (2000b). A nonparametric “trim and fill” method of accounting for publication bias in meta-analysis. Journal of the American Statistical Association, 95(449), 89–98. https://doi.org/10.1080/01621459.200.10473905.
Ebata, A., & Moos, R. (1991). Coping and adjustment in distressed and healthy
adolescents. Journal of Applied Developmental Psychology, 12(1), 33–54.
https://doi.org/10.1016/0193-3973(91)90029-4
Egger, M., Davey Smith, G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical Research Ed.), 315(7109), 629–634.
Eisenberg, N., & Miller, P. (1987). The Relation of Empathy to Prosocial and Related Behaviors. Psychological Bulletin, 101(1), 91–119. https://doi.org/10.1037/0033 -2909.101.1.91
Elison, F., Lennon, R. & Pulos, S. (2006). Investigating the compass of shame: the development of the compass of shame scale. Social behaviour and personality 34: 221-238.
Field, H. (2001). Truth and the Absence of Fact. Oxford: Oxford University Fraley, R. C. (2002). Attachment stability from infancy to adulthood:
Meta-analysis and dynamic modeling of developmental mechanisms. Personality and Social Psychology Review, 6, 123–151. https ://doi.org/10.1207/s1532 7957p spr06 02_03.
George, C., & Solomon, J. (2008). The caregiving system: A behavioral systems approach to parenting. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 833–856). New York, NY: Guilford Press.
Gillath, O., Shaver, P. R., & Mikulincer, M. (2005). An attachment-theoretical approach to compassion and altruism. In P. Gilbert (Ed.), Compassion:
Conceptualizations, research and use in psychotherapy (pp. 121–147). New York: Routledge.
Gladstein, G. (1983). Understanding empathy: Integrating counseling, developmental, and social psychology perspectives. Journal of Counseling Psychology, 30(4), 467–482. https://doi.org/10.1037/0022-0167.30.4.467
Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2, 271–299.
*Gross, C., & Hansen, N. (2000). Clarifying the experience of shame: the role of attachment style, gender, and investment in relatedness. Personality and Individual
Differences, 28(5), 897–907. https://doi.org/10.1016/S0191-8869(99)00148-8 Haidt, J. (2003). The moral emotions. In R. J. Davidson, K. R. Scherer, & H. H.
Goldsmith (Eds.), Handbook of affective sciences (pp. 852–870). Oxford: Oxford University Press.
Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511–524.
Hoeve, M., Stams, G. J. J., Van der Put, C. E., Dubas, J. S., Van der Laan, P. H., & Gerris, J. R. (2012). A meta-analysis of attachment to parents and delinquency. Journal of
abnormal child psychology, 40, 771-785. https://doi.org/10.1007/s10802-011-9608 -1
Hoeve, M., Stams, G. J., Keijsers, L., & Eichelsheim, V. (in press). Hoofdstuk 6: Opvoeding, Gehechtheid, gezinsrelaties en delinquentie. Boom Lemniscaat.
Hoffman, M. L. (2000). Empathy and Moral Development: Implications for Caring and Justice. Cambridge: Cambridge University Press. https//: doi.org/10.1017
/CBO9780511805851
Holmqvist, R. (2008). Psychopathy and affect consciousness in young criminal offenders. Journal of Interpersonal Violence, 23, 209–224. https://doi.org/10.1177
/0886260507309341
Howe, D. (2013). Empathy: What it is and why it matters. New York, New York: Palgrave Macmillan.
Howe, D. (2011) Attachment Across the Life course: A Brief Introduction. Red Globe Press Jenkins, L., Demaray, M., Fredrick, S., & Summers, K. (2016). Associations Among Middle
School Students’ Bullying Roles and Social Skills. Journal of School Violence, 15(3), 259–278. https://doi.org/10.1080/15388220.2014.986675
Joireman, J. A., Needham, T. L., & Cummings, A.-L. (2001). Relationships between dimensions of attachment and empathy. North American Journal of Psychology, 3, 63–80.
*Joireman, J.A., Needham, T., Cummings, A. (2002). Relationships between dimensions of attachment and empathy. North American Journal of Psychology, 4, 63–80
Kestenbaum, R., Farber, E. A., & Sroufe, L. A. (1989). Individual differences in empathy among preschoolers: Relation to attachment history. In N. Eisenberg (Ed.) New directions in child development: No. 44. Empathy and related emotional responses (pp. 51-64). San Francisco: Jossey-Bass.
Kristjánsson, K. (2010). The self and its emotions. Cambridge: Cambridge University Press. Koster, L. (2020) The Association between Attachment and Moral Judgment [Unpublished
masterthesis]. University of Amsterdam
*Laible, D. (2007). Attachment with parents and peers in late adolescence: Links with emotional competence and social behavior. Personality and Individual
Differences, 43(5), 1185–1197. https://doi.org/10.1016/j.paid.2007.03.010
Laible, D., & Thompson, R. (2000). Mother - Child Discourse, Attachment Security, Shared Positive Affect, and Early Conscience Development. Child Development, 71, 1424– 1440. https://doi.org/10.1111/1467-8624.00237
Lefebovre, J., & Krettenauer, T. (2019). Linking Moral Identity With Moral Emotions: A Meta-Analysis. Review of General Psychology, 23(4), 444–457. https://doi.org /10.1177 /1089268019880887
Lewis, H. (1971). Shame and guilt in neurosis. International Universities Press.
Lewis, M. (1997). Altering fate: Why the past does not predict the future. New York: Guilford.
Lewis, M. (2000). Self-conscious emotions: Embarrassment, pride, shame, and guilt. In M. Lichtenberg, J. D., Lachmann, F. M., & Fosshage, J. L. (2011). Psychoanalysis and motiva-
tional systems: A new look. Routledge.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Vol. 49, Sage publications. *Lopez, F. G., Gover, M. R., Leskela, J., Sauer, E. M., Schirmer, L., & Wyssmann, J. (1997).
Personal Relationships, 4, 187–199. https://doi.org/10.1111/j.1475-6811.1997.tb00138.x
Lyons-Ruth, K., & Jacobvitz, D. (1999). Attachment disorganization: Unresolved loss,
relational violence, and lapses in behavioral and attentional strategies. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 520–554). The Guilford Press.
Madigan, S., Brumariu, L., Villani, V., Atkinson, L., & Lyons-Ruth, K. (2016).
Representational and Questionnaire Measures of Attachment: A Meta-Analysis of Relations to Child Internalizing and Externalizing Problems. Psychological
Bulletin, 142(4), 367–399. https://doi.org/10.1037/bul0000029
Main, M., Goldwyn, R. & Hesse, E. (1994). Adult attachment scoring and classification systems: Manual in draft (Version 7.2). Unpublished manuscript, University of California at Berkeley.
Main, M., Kaplan, N., & Cassidy, J. (1985). Security of infancy, childhood, and adulthood: A move to the level of representation. In I. Bretherton & E. Waters (Eds.), Growing points of attachment theory and research. Monographs of the Society for Research in Child Development, 50(1–2, Serial No. 209), 66–106.
Main, M., & Salomon, J. (1990). Procedures for identifying infants classified as
disorganized/disoriented during the Ainsworth strange situation. In M.T. Greenburg, D. Cicchett, & E.M. Cummings (Eds.), Attachment in the preschool years: Theory,
research, and intervention (pp. 121-160).
Malti, T., & Dys, S. P. (2015). A developmental perspective on moral emotions. Topoi: An International Review of Philosophy, 34, 453–459. https://doi.org/10.1007/s11245-013 -9224-5
Manassis, K. (2001). Child–parent relations: Attachment and anxiety disorders. New York: Cambridge University Press.
Mehrabian, A., Young, A., & Sato, S. (1988). Emotional empathy and associated individual
differences. Current Psychology, 7(3), 221–240. https://doi.org/10.1007/BF02686670.
Meissner, W. (2009). Toward a Neuropsychological Reconstruction of Projective
Identification. Journal of the American Psychoanalytic Association, 57(1), 95–129. https://doi.org/10.1177/0003065108329564
Mikulincer, M., & Shaver, P. (2005). Attachment Security, Compassion, and Altruism. Current Directions in Psychological Science, 14(1), 34–38. https://doi.org/10.1111/j.0963-7214.2005.00330.x
Mikulincer, M., & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamics, and change. New York: Guilford.
Mikulincer, M., Florian, V., & Weller, A. (1993). Attachment Styles, Coping Strategies, and Posttraumatic Psychological Distress: The Impact of the Gulf War in Israel. Journal of Personality and Social Psychology, 64(5), 817–826. https://doi.org/10.1037/0022 -3514.64.5.817
Mosquera, P., Manstead, A., & Fischer, A. (2002). The role of honour concerns in emotional reactions to offences. Cognition and Emotion, 16(1), 143–163.
https://doi.org/10.1080/02699930143000167
*Muris, P., Meesters, C., Cima, M., Verhagen, M., Brochard, N., Sanders, A., Kempener, C., Beurskens, J., & Meesters, V. (2014). Bound to feel bad about oneself: Relations between attachment and the self-conscious emotions of guilt and shame in children and adolescents. Journal of Child and Family Studies, 23, 1278–1288. https://doi.org /10.1007/s10826-013-9817-z
