Shame, guilt, and posttraumatic stress symptoms: A three-level meta-analysis
Introduction
Posttraumatic stress disorder (PTSD) is a common psychological disorder with a cross-national lifetime prevalence of 5.6 % after traumatic exposure (Koenen et al., 2017). Since PTSD was first introduced in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-3; American Psychiatric Association [APA], 1980), it has long been categorized as an anxiety disorder. Shame and guilt, though central features observed in shell shock, war neuroses, and combat stress (e.g., Haley, 1974; Ludwig, 1947) and described in rape trauma syndrome (e.g., Burgess, 1983; Dahl, 1989), are largely overlooked until the publication of the DSM-5 (APA, 2013), where shame and guilt are explicitly mentioned for the first time in the diagnostic criteria for PTSD.
However, the associations of shame and guilt with posttraumatic stress symptoms (PTSS) are not always consistent in the extant literature, and may be further complicated by the strong overlap between shame and guilt measures (Cunningham, 2020; Tignor & Colvin, 2019). In addition, the current unprecedented Coronavirus Disease 2019 (COVID-19) pandemic may trigger strong traumatic reactions in some special groups such as survivors of this disease and the front-line key workers (Williamson et al., 2020; Xiao et al., 2020), and the highly distressing and cumulative nature of COVID-19 related stressors may also be a perfect storm to elicit shame and guilt responses (Haller et al., 2020). Therefore, a meta-analysis is warranted to help summarize the magnitude of the associations of shame and guilt with PTSS and explore the potential moderators, which may be informative for the prevention and intervention of COVID-19 related stress symptoms.
Both shame and guilt are self-conscious emotions requiring self-awareness and self-representations (Tracy & Robins, 2004), but they are differentiated in the focus of negative evaluation, affective experience, and behavioral tendency. First, in shame, the focus of negative evaluation is on the entire self (e.g., I am an irresponsible person; Lewis, 1971; Tangney et al., 2011), while in guilt, the focus is more on one’s behaviors (e.g., I didn’t do this seriously; Lewis, 1971; Tangney et al., 2011). Second, shame is more painful than guilt because one’s core self is at stake (Tangney et al., 2007) and is typically accompanied by a sense of worthlessness, incompetence, and inferiority (Bastin et al., 2016; Tangney et al., 2011), while guilt usually generates tension, regret, and remorse for doing something bad (Tangney et al., 2007). Third, people experiencing shame usually attempt to deny, hide, and withdraw physically and/or psychologically from the situation (Tangney et al., 2007; Tignor & Colvin, 2019), while people feeling guilty are often motivated to make up for mistakes by taking reparative actions, such as confession and apology (Bastin et al., 2016; Lawrence & Taft, 2013).
Several theoretical models have depicted the relations of shame and guilt to PTSS. The contemporary cognitive model posits that two key processes, excessively negative appraisals of traumatic events and/or their consequences and disorganized autobiographical memory, produce a sense of serious current threat, which leads to persistent PTSD (Ehlers & Clark, 2000). Evaluations concerning violation of individuals’ important internal standards may elicit threat to the self and then shame, while evaluations regarding personal responsibility for a traumatic event and/or its consequences may lead to guilt (Ehlers & Clark, 2000; Lawrence & Taft, 2013). Another clinical model of shame- and guilt-based PTSD proposed by Lee et al. (2001) indicates that shame may arise from the congruence of the traumatic event’ meaning with underlying shame schemas that have deeper implications for the self and/or others, or from the breakdown of a positive self-identity, while guilt may come up when the meaning of the traumatic event represents a deviation or violation from behavioral standards and/or a sense of responsibility for causing harm to others. Also, PTSD may emerge from the impacts of the dysregulation of emotions (e.g., shame and guilt) on social cognition and interpersonal functions (Budden, 2009; Sharp et al., 2012).
Although the two emotions (i.e., shame and guilt) and PTSS are theoretically related, the empirical findings on the associations of shame and guilt with PTSS are not always consistent. For the relation between shame and PTSS, a positive correlation is reported in the vast majority of studies (e.g., Dewey et al., 2014; Feiring et al., 2002a, 2002b; Feiring & Taska, 2005; Sippel & Marshall, 2011) and nonsignificant correlation in a few studies (e.g., Barr, 2012; Ojserkis et al., 2014; Shin et al., 2014; Straub et al., 2018). The findings on the relation between guilt and PTSS are even more mixed, ranging from a positive correlation, to no correlation, to a negative correlation (see Pugh et al., 2015 for a review). These highly heterogeneous findings on the relations of shame and guilt to PTSS suggest a need to explore the factors that might explain this heterogeneity.
In the following section, we will briefly discuss the potential moderators that possibly have theoretical meanings and are included in this meta-analysis.
As Cunningham (2020) said, comparing studies of guilt and shame in trauma-exposed samples seems difficult, because some instruments measure shame or guilt proneness, whereas others tap into trauma-specific emotions. The work of Kim et al. (2011) has demonstrated a significant variation in weighted effect sizes of the association between sham/guilt and depressive symptoms across studies using different types of scales. The study by Cândea and Szentagotai-Tătar (2018) also found type of measurement significantly moderates the effect size of the correlations of shame and guilt with some anxiety symptoms. Based on the above two meta-analytic studies, we classified shame measures into three subcategories, assessing generalized shame, contextual shame, and trauma-specific shame, respectively, and divided guilt scales into generalized guilt, contextual guilt, and trauma-specific guilt scales. Generalized shame and guilt scales separate the experiences of shame and guilt from particular contexts and do not tap specific elements (e.g., phenomenological, behavioral, motivational) of the two emotions, such as the adjective lists requiring respondents to report the frequency of these emotional experiences and the extent to which the adjectives describe themselves (e.g., Personal Feelings Questionnaire-2, PFQ-2; Harder & Zalma, 1990). In contrast, contextual shame and guilt measures embed the emotional experiences in particular scenarios, or tap specific elements of the two emotions and the situations in which they arise. For example, the Test of Self-Conscious Affect-3 (TOSCA-3; Tangney et al., 2000) contains shame- and guilt-relevant scenarios and asks respondents to indicate the likelihood of reactions to these situations. The Internalized Shame Scale (ISS; Cook, 1994), though not invoking any particular scenario, is sensitive to specific phenomenological elements of shame the respondents experience in a certain context (e.g., “I could beat myself over the head club when I make a mistake”). Additionally, as PTSS is a trauma-specific construct in our study, we divorced the traumatic context from the general context, and obtained trauma-specific shame and guilt measures which assess the shame- and guilt-relevant responses to traumatic events (e.g., Trauma-Related Guilt Inventory, TRGI; Kubany et al., 1996). As trauma-specific measures are closely tied to traumatic situations, we expected the strongest links of trauma-specific shame and guilt to PTSS compared to generalized or contextual shame and guilt.
Previous research has revealed cross-cultural differences in shame and guilt. First, the difference between shame and guilt seems less pronounced in Eastern culture than in Western culture (Wong & Tsai, 2007). A study by Bedford (2004) identified four types of shame (diu lian, can kui, xiu kui, and xiu chi) and three types of guilt (nei jiu, zui e gan, and fan zui gan) in Chinese which are not distinguishable from each other in English, and found that xiu kui, a strong feeling of shame in Chinese, resembled guilt in English in that they are both not related to others’ judgments. Moreover, some types of Chinese shame such as can kui and xiu kui serve to promote self-improvement (Bedford & Hwang, 2003), making them more similar to, rather than different from, English guilt. Second, Eastern culture usually considers shame and guilt more valuable and constructive than does Western culture (Wong & Tsai, 2007; Yang et al., 2012). Individuals in Eastern countries are encouraged to view the self in connection with others (Markus & Kitayama, 1991), and others’ thoughts and feelings are as important and meaningful as one’s own (Wong & Tsai, 2007). In this context, shame and guilt involving situational changes in one’s self or behaviors are viewed as normative, explicitly valued and expected (Cho, 2000). By contrast, undesirable guilt is more frequent in Western countries than in Eastern countries (Eid & Diener, 2001), and shame is particularly damning in a Western context that values personal achievement (Sheikh, 2014). Then, we would explore whether the culture type moderates the relations of shame and guilt to PTSS.
Studies consistently show stronger associations between trauma-related guilt and PTSS among military samples than among traumatized civilians (Pugh et al., 2015). In addition, shame has been related to human-generated trauma which contains interpersonal elements of domination and conquest (Kallstrom-Fuqua et al., 2004). Using a sample of college students, the study by La Bash and Papa (2014) found that interpersonal trauma was related to higher levels of peritraumatic shame and PTSS than noninterpersonal trauma. As such, the varying nature of traumatic stressors and samples may influence the relationships of shame and guilt with PTSS, and we would examine the possible moderating effects of these variables.
Age and gender would be included in moderator analysis, because they were found to be relevant for PTSD (Schäfer et al., 2019), and preliminary evidence showed that age negatively predicted the association of guilt with PTSD and the percentage of females also negatively predicted the associations of shame and guilt with PTSD (Cândea & Szentagotai-Tătar, 2018).
As theoretical models of PTSD recognize the critical roles of shame and guilt and empirical findings are controversial, a next and vital step is to use a quantitative review method to examine the relationships of shame and guilt to PTSS. A recent meta-analysis by López-Castro et al. (2019) revealed a significant moderate correlation between shame and PTSD, but they did not consider the possible confounding effect of guilt on this effect size. Another meta-analysis by Cândea and Szentagotai-Tătar (2018) controlled for the effect of guilt or shame, but mainly focused on the associations of shame and guilt with anxiety symptoms, in which PTSD was still categorized as an anxiety disorder and trauma-specific variables were not explored as potential moderators. Besides, the two meta-analyses have two other limits. One is the traditional approach (i.e., calculating the average) they used to handle the dependent effect sizes, which can lead to a loss of some information in effect sizes and thus less precise parameter estimates and a lower statistical power (Cheung, 2014). The other is the neglect of the possible cultural effects discussed above. Thus, the first purpose of this study was to use a three-level meta-analysis for the first time to quantify the strength of the associations of shame and guilt with PTSS. Also, we aimed to examine the unique association of shame with PTSS by controlling for the effect of guilt, and the unique relationship between guilt and PTSS by partialling out the effect of shame. Finally, we would test several possible moderators to enhance our understanding of the relations of shame and guilt to PTSS.
Section snippets
Method
This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al., 2009) and has been registered at PROSPERO (CRD42020189605).
Results
Overall, 325 effect sizes were computed from 92 studies (89 papers) included in the meta-analysis (88 for the bivariate correlation between shame and PTSS, 193 for the bivariate correlation between guilt and PTSS, 22 for the partial correlation between shame and PTSS, and 22 for the partial correlation between guilt and PTSS). Our search was updated to November 13 2020, and the studies included in the meta-analysis were published between 1995 and 2020. There are four eastern studies; two of
Discussion
The current meta-analyses estimated the strength and significance of the relationships of shame and guilt with PTSS, using the full bivariate and partial (controlling for the effect of shame or guilt) correlation data, respectively. We also examined potential moderators of the magnitude of these relationships (i.e., culture, type of sample, type of trauma, type of shame and guilt measure, age, and gender).
Conclusion
The current study summarized the extant literature on the associations of shame and guilt with PTSS, and demonstrated the significantly positive correlations of shame and guilt with PTSS, as well as the moderating role of shame and guilt measure type. In addition, the culture tended to moderate the relation between guilt and PTSS. We believed that clarifying the conceptual distinctions between shame and guilt and their associated maladaptive characteristics would contribute to a better
Funding
This work was supported by the self-determined research funds of Central China Normal University from the colleges’ basic research and operation of Ministry of Education of China (grant number: CCNU20TD001); the Key Program of Institute of Wuhan Studies of Jianghan University (grant number: IWHS20201007).
Declaration of Competing Interest
The authors declare that they have no conflicts of interest.
Acknowledgement
None.
References (149)
- et al.
Feelings of shame, embarrassment and guilt and their neural correlates: A systematic review
Neuroscience & Biobehavioral Reviews
(2016) Multivariate meta-analysis
- et al.
Psychometric properties of the PTSD Checklist (PCL)
Behaviour Research and Therapy
(1996) The role of shame in posttraumatic stress disorder: A proposal for a socio-emotional model for DSM-V
Social Science & Medicine
(2009)- et al.
Shame-proneness, guilt-proneness and anxiety symptoms: A meta-analysis
Journal of Anxiety Disorders
(2018) Shame and guilt in PTSD
- et al.
A prospective investigation of the role of cognitive factors in persistent posttraumatic stress disorder (PTSD) after physical or sexual assault
Behaviour Research and Therapy
(2001) - et al.
A cognitive model of posttraumatic stress disorder
Behaviour Research and Therapy
(2000) - et al.
Psychological trauma and PTSD symptoms as predictors of substance dependence treatment outcomes
Behaviour Research and Therapy
(2007) - et al.
Emotion regulation therapy for generalized anxiety disorder
Cognitive and Behavioral Practice
(2013)
An exploration of shame measures—I: The other as Shamer scale
Personality and Individual Differences
Treating PTSD in suicidal and self-injuring women with borderline personality disorder: Development and preliminary evaluation of dialectical behavioral therapy prolonged exposure protocol
Behavior Research and Therapy
An abbreviated PTSD checklist for use as a screening instrument in primary care
Behavior Research and Therapy
Shame, posttraumatic stress disorder, and intimate partner violence perpetration
Aggression and Violent Behavior
Trauma informed guilt reduction therapy with combat veterans
Cognitive and Behavioral Practice
Investigating the relationship between guilt and shame proneness and moral injury in veterans that have experienced active deployment
Military Behavioral Health
Diagnostic and statistical manual of mental disorders
Diagnostic and statistical manual of mental disorders
Fitting three-level meta-analytic models in R: A step-by-step tutorial
The Quantitative Methods for Psychology
Predicting depressive symptoms with a new measure of shame: The Experience of Shame Scale
British Journal of Clinical Psychology
Risk factors for child sexual abuse victimization: A meta-analytic review
Psychological Bulletin
Shame proneness, negative cognitions, and posttraumatic stress among women with a history of sexual trauma
Journal of Aggression, Maltreatment & Trauma
Less is more: Meta-analyses of sensitivity and attachment interventions in early childhood
Psychological Bulletin
A dyadic analysis of negative emotion personality predisposition effects with psychological distress in neonatal intensive care unit parents
Psychological Trauma: Theory, Research, Practice, and Policy
The individual experience of guilt and shame in Chinese culture
Culture & Psychology
Guilt and shame in Chinese culture: A cross-cultural frame-work from the perspective of morality and identity
Journal for the Theory of Social Behaviour
Cross-cultural psychology: Research and applications
The development of a Clinician-Administered PTSD Scale
Journal of Traumatic Stress
Measuring health status among survivors of burn injury: revisions of the Burn Specific Health Scale
The Journal of Trauma
Traumatic symptom inventory professional manual
Psychometric review of the Trauma Symptoms Checklist-40
Detailed assessment of posttraumatic stress
Rape trauma syndrome
Behavioral Sciences & the Law
Applied meta-analysis for social science research
Modeling dependent effect sizes with three-level meta-analyses: A structural equation modeling approach
Psychological Methods
Meta-analyzing dependent correlations: An SPSS macro and an R script
Behavior Research Methods
Public opinion as personal cultivation: A normative notion and a source of social control in traditional China
International Journal of Public Opinion Research
A power primer
Psychological Bulletin
Introducing the GASP scale: A new measure of guilt and shame proneness
Journal of Personality and Social Psychology
Internalized shame scale: Professional manual
Residuals and influence in regression
A relative weights comparison of trauma‐related shame and guilt as predictors of DSM‐5 posttraumatic stress disorder symptom severity among US veterans and military members
British Journal of Clinical Psychology
Acute response to rape-a PTSD variant
Acta Psychiatrica Scandinavica
Assessment of a new self-rating scale for post-traumatic stress disorder
Psychological Medicine
Posttraumatic growth and posttraumatic distress: A longitudinal study
Psychological Trauma: Theory, Research, Practice, and Policy
Links between specific posttrauma appraisals and three forms of trauma-related distress
Psychological Trauma: Theory, Research, Practice, and Policy
Development of the trauma appraisal questionnaire
Journal of Aggression, Maltreatment & Trauma
Do peritraumatic emotions differentially predict PTSD symptom clusters? Initial evidence for emotion specificity
Psychological Reports
Norms for experiencing emotions in different cultures: Inter- and intranational differences
Journal of Personality and Social Psychology
The modified PTSD Symptom Scale: A brief self-report measure of post-traumatic stress disorder
Behavior Therapist
Cited by (23)
Non-fear emotions in changes in posttraumatic stress disorder symptoms during treatment
2024, Journal of Behavior Therapy and Experimental PsychiatryA network analysis exploration of PTSD, depression, somatic complaints, and anxiety disorders in a Colombian sample of victims of the Armed conflict
2024, Journal of Affective Disorders ReportsEffects of digital psychological interventions on physical symptoms in cancer patients: A systematic review and meta-analysis
2023, General Hospital PsychiatryCardiothoracic surgeons as second victims: We, too, are at risk
2023, Journal of Thoracic and Cardiovascular SurgeryLaboratory models of post-traumatic stress disorder: The elusive bridge to translation
2022, NeuronCitation Excerpt :There also remains a debate on whether the term “fear” is too subjective to be applied to paradigms that evaluate the defensive response systems of nonhuman animals (LeDoux, 2020; Mobbs et al., 2019). Finally, the plethora of complex human emotional responses to trauma, including guilt, shame, and moral injury (Shi et al., 2021), likewise render the fear conditioning model of PTSD an incomplete recapitulation of the human trauma response. Defensive behavior is an ordinary response to stimuli that reliably predict danger.