Associations Between Generalized Shame and Emotional Processing Difficulties Among Individuals Seeking Mental Health Care
Alicia Spidel, PhD & David Kealy, Ph.D.
January 20, 2025
Clinicians interested in supporting individuals with emotional processing difficulties play a critical role in fostering wellbeing, as emotional experiences deeply influence cognitive, behavioral, and social functioning. For those in psychotherapy, understanding the mechanisms of emotional processing is essential as it involves integrating unpleasant or disturbing emotions to maintain focus, adapt to varying experiences, and uphold behavioral flexibility (Baker et al., 2010). Effective emotional processing mitigates disruptions in behavior and adaptation when faced with challenging emotional experiences. Impairments in emotional processing, such as limited awareness of feelings, avoidance, suppression, or maladaptive emotional expression, can significantly hinder a client’s ability to engage in therapeutic work. Addressing these challenges within clinical settings is crucial for promoting emotional resilience and improving therapeutic outcomes and maladaptive or inappropriate emotional expression (Baker et al., 2010).
In some instances, maladaptive actions may be taken to deal with challenging feelings in the absence of constructive emotional processing, including substance misuse, aggression or self-harm, and risk-taking behaviour. Constructs pertaining to impaired emotional processing such as alexithymia––difficulty identifying and expressing feelings––and emotion dysregulation have been identified as primary vulnerabilities underlying a range of mental health complaints and disorders (Kealy et al., 2017). Individuals who have difficulty with emotional processing also tend to report more interpersonal problems (Wei et al., 2005) and have interpersonal relationships with diminished quality (Humphreys et al., 2009). Furthermore, emotional processing difficulties may increase risk for suicidal ideation and behaviour (Iskric et al., 2020). Given these implications of emotional processing difficulties, it is important to identify associated individual difference variables as intervention efforts may require targeting of such features to effect changes in emotional processing.
One individual difference variable that may be implicated in impaired emotional processing is generalized shame. While shame can be a transient affective state that arises in the context of perceived transgressions, involving attribution to the self––not only one’s behaviour––and feelings of inadequacy or humiliation (Elshout et al., 2017), it can also persist as a generalized and chronic tendency to feel that one is unacceptable and bad (Dolezal, 2022). In contrast to guilt, which may prompt reflection and interpersonal repair (Joireman, 2004), shame is linked with apprehension about close interpersonal connections (Wu et al., 2020) and with self-rumination rather than reflection (Joireman, 2004). Indeed, generalized shame may inhibit constructively focusing on, let alone seeking support for, challenging emotional experiences. Thus, a high level of generalized shame could be associated with aspects of impaired emotional processing.
Interestingly, limited research has examined shame in relation to various forms of emotional processing difficulties. While some research has shown associations between shame and alexithymia (Rice et al., 2020) and emotion suppression (Velotti et al., 2017), limited study has occurred among clinical samples encompassing individuals who may have significant concerns about emotional processing along with elevated psychological distress. Indeed, because few of these studies have controlled for depressive symptoms––often entangled with shame and impaired emotional functioning––it remains unclear whether generalized shame contributes uniquely to emotional processing difficulties. The present study was thus aimed at examining associations between generalized shame and several forms of impaired emotional processing among outpatients seeking mental health care, investigating whether variance in emotional processing is accounted for by shame after controlling for depressive symptoms.
Method
Participants were consecutively admitted adult patients attending a publicly-funded outpatient mental health clinic in the Greater Vancouver area of Canada. Clinic care consists of short-term psychotherapy by masters-level clinicians with occasional psychiatric consultation. Clinical interviews are used to determine patient eligibility for service on the basis of serious and impairing mood- or anxiety-related mental health disorder. Comprehensive case formulation, rather than psychiatric diagnosis, guides care; hence, diagnoses were not available in the data. Institutional review boards granted ethics approval for the study.
Informed consent for the study was obtained from 100 patients. However, one participant did not complete assessment questionnaires leaving a sample of 99 patients for the present study. The study procedure consisted of the completion of assessment questionnaires upon commencing clinic services. Participants had a mean age of 36.07 years (SD=11.88; range=18-61). The majority were female at 68 percent. Regarding ethnicity, 67.7% identified as Caucasian, 9.1% as South Asian, 8.1% as First Nations, 6.1% as Asian, and 9% as other ethnicities. Relationship status included 42.4% in spousal relationships and 40.4% single with 16.2% had been separated, divorced, or widowed.
More than half of the participants (52.5%) reported some form of post-secondary education. Regarding employment, 22.2% and 17.2% were employed full-time and part-time, respectively; 35.4% were unemployed and 7.1% were receiving disability benefits with the remaining 17.2% not seeking employment as either stay-home parents, students, or retirees. The majority of participants (63.6%) reported previous psychiatric treatment.
Shame was assessed using the shame subscale of the Personal Feelings Questionnaire-2 (PFQ-2; Harder et al., 1993). The PFQ-2 shame subscale consists of ten adjective-based items referring to affective experiences comprising generalized shame scored from 0 (never) to 4 (continuously or almost continuously). Impaired emotional processing was assessed using the Emotional Processing Questionnaire (EPS-25; Baker et al., 2010). The EPS-25 is a 25-item self- report inventory comprised of five subscales that correspond to different types of emotional processing difficulties. Items refer to responses of emotional experiences over the past week rated from 0 (completely disagree) to 9 (completely agree). Five subscales refer to emotional control (avoidance, suppression), difficulty discerning emotions (impoverished emotional experience), emotional dysregulation (unregulated emotion), and persistent, intrusive emotions (signs of unprocessed emotion). A total score indicates overall impairment in emotional processing. Depressive symptoms were assessed using the self-report Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., 2001) consisting of nine items referring to frequency of major depressive disorder symptoms over the past two weeks. Items are scored from 0 (not at all) to 3 (nearly every day), with higher scores indicating more severe depressive symptoms.
Acceptable internal consistency was observed for all measures indicated by coefficients displayed in Table 1.
Table 1. Descriptive statistics, internal consistency coefficients, and correlations between shame, depressive symptoms, and impaired emotional processing domains.
| M (SD) | a | Generalized shame | Depressive symptoms | |
|---|---|---|---|---|
| Generalized shame | 20.70 (6.66) | .82 | — | — |
| Depressive symptoms | 17.85 (6.21) | .88 | .36*** | — |
| EPS total | 5.33 (1.41) | .90 | .40*** | .38*** |
| Suppression | 5.34 (2.03) | .87 | .30** | .20 |
| Signs of unprocessed emotion | 6.40 (1.83) | .83 | .37*** | .41*** |
| Unregulated emotion | 4.96 (1.87) | .72 | .36*** | .29** |
| Avoidance | 5.15 (1.68) | .69 | .28** | .19 |
| Impoverished emotional experience | 4.81 (1.88) | .75 | .20* | .36*** |
*p < .05; **p < .01; ***p < .001
Analyses, performed using SPSS 25, consisted of zero-order correlations among study variables––including tests of age and gender as potential confounders––followed by a series of linear regression models. Each linear regression included depressive symptoms and shame as predictors, with impaired emotional processing responses as criterion variables. The EPS-25 total score was examined first, followed by each EPS- 25 subscale in separate regression models.
Results
Descriptive statistics, internal consistency coefficients, and zero-order correlations are presented in Table 1. Acceptable internal consistency was observed for all study measures. Generalized shame was significantly positively associated with the total EPS-25 score and with each of the five types of impairment in emotional processing. Tests of age and gender as potential confounders (zero-order correlations) revealed no significant associations with the variables of interest. Regression analyses thus proceeded with the EPS-25 total score and each subscale as criterion variables, examining generalized shame as a predictor after controlling for depressive symptoms. Standardized coefficients and changes in R2 are presented in Table 2.
Table 2. Linear regression models examining generalized shame and depressive symptoms as in relation to impaired emotional processing domains.
| Criterion variable | Predictors | DR2 | β | t | p |
|---|---|---|---|---|---|
| EPS-25 total score | Depressive symptoms | .15 | .27 | 2.79 | .006 |
| Generalized shame | .08 | .29 | 3.03 | .003 | |
| Suppression | Depressive symptoms | .04 | .10 | .95 | .345 |
| Generalized shame | .06 | .26 | 2.49 | .015 | |
| Signs of unprocessed emotion | Depressive symptoms | .17 | .32 | 3.29 | .001 |
| Generalized shame | .06 | .25 | 2.63 | .010 | |
| Unregulated emotion | Depressive symptoms | .09 | .18 | 1.82 | .072 |
| Generalized shame | .07 | .29 | 2.84 | .005 | |
| Avoidance | Depressive symptoms | .04 | .10 | .96 | .341 |
| Generalized shame | .05 | .24 | 2.28 | .025 | |
| Impoverished emotional experience | Depressive symptoms | .13 | .33 | 3.22 | .002 |
| Generalized shame | .01 | .08 | .75 | .454 | |
| Note: boldface indicates statistical significance |
Generalized shame was found to contribute independently to the variance in overall impairment in emotional processing and to each facet represented by EPS-25 subscales, with the exception of impoverished emotional experience. Thus, beyond the effects of depressive symptoms, generalized shame was significantly associated with avoidance, suppression, and dyscontrol of emotional experience, as well as the persistence of intrusive negative feelings accounting for between 5-7% of the variance in these emotional processing domains.
Discussion
Findings from the present study indicate that among individuals seeking mental health treatment, generalized shame is robustly associated with impaired emotional processing. Indeed, after controlling for the effects of depressive symptoms, shame accounted for unique variance in four out of five emotional processing domains: suppression, avoidance, signs of unprocessed emotion, and unregulated emotion.
These findings suggest the possibility of trait-like susceptibility to shame being a contributing factor in emotional processing difficulties involving both the inhibition of and reactivity to emotional experience. However, given the cross-sectional, correlational nature of these findings obtained in a relatively small sample, such inference should be considered preliminary pending further investigation.
The perception of the self as bad or inadequate involved in generalized shame may lead some individuals to conclude that the experience of negative emotion is a confirmation of one’s badness. This could lead some shame-prone individuals to employ excessive avoidance or suppression to keep difficult feelings from awareness in order to mitigate exacerbation of shame. In other words, these individuals may lack a compassionate stance toward the presence of negative feelings, thereby reducing efforts to understand and address their emotions. High generalized shame is also associated with adverse childhood experience (Wojcik et al., 2019). Some forms of childhood adversity, such as chronic emotional abuse and neglect, may reduce developmental practicing of emotional processing both intrapersonally, in the form of self-reflection and reappraisal, and interpersonally, through disclosure and discussion of feelings among supportive others. Similarly, systemic violence and oppression can create shaming experiences and lead to internalized shame among individuals in marginalized groups (Nadal et al., 2021). Further research is needed to examine shame as a potential mediator of the effects of adversity, discrimination, or violence on emotional processing among vulnerable populations.
The position of shame with respect to social approach and avoidance is complex; induced shame may motivate social approach to repair a threatened self, though if repair is unavailable, shame can motivate withdrawal to protect the self (de Hooge et al., 2010). It is possible that generalized shame associated with a pessimistic outlook (Kealy et al., 2022) might more frequently invoke social withdrawal, thereby reducing exposure to interpersonal interactions that down-regulate negative emotion (Zaki & Williams, 2013). The result may be unprocessed emotion that overwhelms the individual’s coping responses or becomes channeled into maladaptive behaviours. Further research is needed to examine mediators and moderators, including dispositional, developmental, and social factors, of the associations observed between shame and domains of impaired emotional processing.
Several limitations must be noted regarding this preliminary study. First, the sample was relatively small and based entirely at a single Canadian clinic. While a larger sample might reveal stronger effects in the associations we observed, the degree to which these findings might replicate across other clinical populations is unclear. Second, given the cross-sectional nature of the data, directionality cannot be inferred. Furthermore, assessment was restricted to self-report, introducing the possibility of bias that could be mitigated by observer-rated or task- or scenario-based measurement.
Despite these limitations, the present study indicates generalized shame as a potential impediment to enhanced emotional processing, even after accounting for depressive symptoms. This raises the possibility that clinical attention to reducing depression may not be sufficient for helping patients achieve greater facility for emotional processing.
There are some clinical implications of these findings. The experience of shame may influence some clients to drop out of therapy early (Foa et al., 2000). Therefore, recognizing shame and understanding how this may impact emotional processing in the context of treatment engagement is essential for clinicians to help clients persist in the emotionally challenging task of psychotherapy. Clinicians may need to pay particular attention to prioritizing an accepting and safe therapeutic relationship to help mitigate the client’s feelings of shame. While many theories consider how to address shame with therapeutic interventions, only two theoretical frameworks were designed specifically for this purpose: Compassion-Focused Therapy (Gilbert, 2009) and Shame Resilience Theory (Brown, 2006). Though relatively limited, research has indicated the potential for these approaches to have positive therapeutic outcomes, especially for clients with high amounts of shame (Craig et al., 2020). While further research is needed to determine whether such approaches can enhance emotional processing abilities, the integration of shame- focused psychotherapeutic interventions may be worthy of consideration in helping clients overcome emotional processing difficulties.