Amy H. Mezulis

Is There a Universal Positivity Bias in Attributions? A Meta-Analytic Review of Individual, Developmental, and Cultural Differences in the Self-Serving Attributional Bias

Researchers have suggested the presence of a self-serving attributional bias, with people making more internal, stable, and global attributions for positive events than for negative events. This study examined the magnitude, ubiquity, and adaptiveness of this bias. The authors conducted a meta-analysis of 266 studies, yielding 503 independent effect sizes. The average d was 0.96, indicating a large bias. The bias was present in nearly all samples. There were significant age differences, with children and older adults displaying the largest biases. Asian samples displayed significantly smaller biases (d = 0.30) than U.S. (d = 1.05) or Western (d = 0.70) samples. Psychopathology was associated with a significantly attenuated bias (d = 0.48) compared with samples without psychopathology (d = 1.28) and community samples (d = 1.08). The bias was smallest for samples with depression (0.21), anxiety (0.46), and attention-deficit/hyperactivity disorder (0.55). Findings confirm that the self-serving attributional bias is pervasive in the general population but demonstrates significant variability across age, culture, and psychopathology.

The developmental origins of cognitive vulnerability to depression: Temperament, parenting, and negative life events in childhood as contributors to negative cognitive style.

Cognitive models of depression have been well supported with adults, but the developmental origins of cognitive vulnerability are not well understood. The authors hypothesized that temperament, parenting, and negative life events in childhood would contribute to the development of cognitive style, with withdrawal negativity and negative parental feedback moderating the effects of negative life events to predict more depressogenic cognitive styles. These constructs were assessed in 289 children and their parents followed longitudinally from infancy to 5th grade; a subsample (n = 120) also participated in a behavioral task in which maternal feedback to child failure was observed. Results indicated that greater withdrawal negativity in interaction with negative life events was associated with more negative cognitive styles. Self-reported maternal anger expression and observed negative maternal feedback to childs failure significantly interacted with childs negative events to predict greater cognitive vulnerability. There was little evidence of paternal parenting predicting child negative cognitive style.

Stress and Emotional Reactivity as Explanations for Gender Differences in Adolescents’ Depressive Symptoms

In this longitudinal study, we examined whether certain types of stressful events and how individuals respond to these events would explain gender differences in depressive symptoms among adolescents. We hypothesized that certain stressful events would mediate the relationship between gender and depressive symptoms. We also hypothesized that individual differences in emotional reactivity would impact part of this relationship. Lastly, we examined whether gender differences in early childhood temperament might explain gender differences in emotional reactivity in adolescence. We examined these hypotheses in a sample of 315 adolescents (51% females; 93% Caucasian; 3% African–American; and 1% each Hispanic, Asian–American, and Native American) participating in a longitudinal study of child development since birth. We used multiple regression and constrained nonlinear regression to analyze the data. Results indicated that stressful events significantly mediated gender differences in depression, and that individual differences in emotional reactivity to these stressors significantly moderated the relationship between stress and depression. We also observed significant gender differences in emotional reactivity to these stressors; temperamental differences in withdrawal negativity in infancy were marginally significant in mediating gender differences in emotional reactivity to stress in adolescence.

Gender Differences in the Cognitive Vulnerability-Stress Model of Depression in the Transition to Adolescence

This study examined whether the cognitive vulnerability-stress model of depression may contribute to our understanding of the gender difference in depression in adolescence. Specifically, we examined emergent gender differences in depressive symptoms, cognitive style, and stress in the context of exposure, cognitive scar, and stress generation models. We also examined whether gender moderated the cognitive vulnerability-stress effects on depression. Participants were 366 youth from a community sample who completed measures of depressive symptoms, stress, and negative cognitive style at ages 11, 13, and 15. Data were analyzed longitudinally using multi-level modeling and structural equation modeling. Results indicated that gender differences in depressive symptoms emerged prior to gender differences in cognitive vulnerability and stressful life events; depressive symptoms significantly mediated the emergent gender difference in cognitive style and dependent interpersonal stress. Gender also moderated several components of the cognitive vulnerability-stress model. Girls showed stronger associations between stress and depression over time, and the cognitive vulnerability-stress interaction was significant in predicting girls’ but not boys’ depression trajectories.

The influence of child gender role and maternal feedback to child stress on the emergence of the gender difference in depressive rumination in adolescence.

Extensive research has linked a greater female tendency to ruminate about depressed feelings or mood to the gender difference in depression. However, the developmental origins of the gender difference in depressive rumination are not well understood. We hypothesized that girls and women may be more likely to ruminate because rumination represents a gender-stereotyped coping style that is associated with a more feminine gender role identity, maternal encouragement of emotion expression, and passive coping responses to stress. This study examined whether child self-reported gender role identity and observed maternal responses to child stress mediated the emergent gender difference in depressive rumination in adolescence. Maternal gender role attitudes were further hypothesized to moderate the relationship between child sex and mediating variables. Rumination and gender role identity were assessed in 316 youths and their mothers in a longitudinal study from age 11 to age 15; in addition, 153 mother-child dyads participated in an observational task at age 11 from which maternal responses to a child stressor were coded. Results indicated that greater feminine gender role identity among children and encouragement of emotion expression by mothers at age 11 significantly mediated the association between child sex and the development of depressive rumination at age 15, even after controlling for rumination at age 11. Maternal gender role attitudes significantly moderated the relationship between child sex and maternal encouragement of emotion expression, such that mothers who endorsed more traditional gender role attitudes themselves were particularly likely to encourage emotion expression in their daughters.

Domain Specificity of Gender Differences in Rumination

Rumination, defined as repetitively focusing on the causes, meanings, and symptoms of depressed mood, has been associated with the onset, maintenance, and exacerbation of depression. Although several studies have identified gender differences in rumination, with females being more likely to ruminate than males, no study to date has examined whether gender differences in rumination vary across domains. The current study examined whether females may be more vulnerable to rumination in response to negative events in domains hypothesized to be particularly salient to females, such as interpersonal or body image domains. To this end, 259 undergraduates (148 females, 111 males) completed three different rumination questionnaires, one assessing rumination in response to “sad or depressed mood,” another assessing rumination in response to “stressful events,” and a third assessing rumination in response to a series of negative events in 3 separate domains: achievement, interpersonal, and body image/attractiveness. On all measures, females reported more rumination than males. The effect sizes for gender differences in rumination in response to depressed mood or stressful events, as well as the effect size for gender differences in rumination in response specifically to achievement events, were small. By contrast, effect sizes for gender differences in rumination in response specifically to negative interpersonal and body image events were in the moderate to large range. Results support the hypothesis that although females display an overall greater tendency than males to ruminate in response to negative events and depressed mood, gender differences in rumination may be particularly pronounced in the interpersonal and body image domains.

Adolescent emotionality and effortful control: Core latent constructs and links to psychopathology and functioning

Temperament is associated with important outcomes in adolescence, including academic and interpersonal functioning and psychopathology. Rothbarts temperament model is among the most well-studied and supported approaches to adolescent temperament, and contains 3 main components: positive emotionality (PE), negative emotionality (NE), and effortful control (EC). However, the latent factor structure of Rothbarts temperament measure for adolescents, the Early Adolescent Temperament Questionnaire Revised (EATQ-R; Ellis & Rothbart, 2001) has not been definitively established. To address this problem and investigate links between adolescent temperament and functioning, we used confirmatory factor analysis to examine the latent constructs of the EATQ-R in a large combined sample. For EC and NE, bifactor models consisting of a common factor plus specific factors for some subfacets of each component fit best, providing a more nuanced understanding of these temperament dimensions. The nature of the PE construct in the EATQ-R is less clear. Models replicated in a hold-out dataset. The common components of high NE and low EC where broadly associated with increased psychopathology symptoms, and poor interpersonal and school functioning, while specific components of NE were further associated with corresponding specific components of psychopathology. Further questioning the construct validity of PE as measured by the EATQ-R, PE factors did not correlate with construct validity measures in a way consistent with theories of PE. Bringing consistency to the way the EATQ-R is modeled and using purer latent variables has the potential to advance the field in understanding links between dimensions of temperament and important outcomes of adolescent development.

Cognitive mechanisms linking low trait positive affect to depressive symptoms: A prospective diary study

Low trait positive affect represents an affective vulnerability to depression, but little research has examined mechanisms linking low trait positive affect to depressive symptoms. The current study investigated whether the cognitive strategies of dampening and positive rumination mediated the prospective association between low trait positive affect and depressive symptoms. Participants were 209 undergraduate students who participated in an eight-week online study. Depressive symptoms and trait temperament were assessed at baseline, followed by seven weekly questionnaires which assessed cognitive strategies in response to weekly events and weekly depressive symptoms. Data were analysed using multilevel modelling following the mediation approach proposed by Nezlek. Results indicated that low trait positive affect significantly predicted less positive rumination but not greater dampening in response to weekly positive events. Less positive rumination in response to weekly positive events partially mediated the association between low trait positive affect and greater depressive symptoms across the study.

Pathways to Depressive Symptoms in Young Adults: Examining Affective, Self-Regulatory, and Cognitive Vulnerability Factors

Recent models of depression (e.g., from Hyde and colleagues) have integrated affective and cognitive vulnerability factors, positing that a temperamental factor (i.e., negative emotionality) contributes to the development of cognitive vulnerability factors, which in turn conveys risk for depressive symptoms. Recent literature suggests that effortful control may reduce the strength of the relationship between affective and cognitive factors. However, few studies have examined the different cognitive vulnerability factors (cognitive style, brooding, and stress-reactive rumination) through which negative emotionality contributes to depressive symptoms, or how effortful control may influence these paths. 315 young adults (72% women, M age = 20.7 yr., SD = 1.4) answered psychometric measures of temperament factors (negative emotionality and effortful control), and three cognitive vulnerability factors (cognitive style, brooding, stress-reactive rumination), and depressive symptoms. Two hypotheses were tested concerning mediation and moderation of the relationship between negative emotionality and depressive symptoms. Cognitive style and brooding were significant mediators of this relationship, while effortful control did not moderate relations between negative emotionality and cognitive vulnerability factors. Results support models that integrate affective and cognitive vulnerability factors predicting depressive symptoms.

Pathways linking temperament and depressive symptoms: a short-term prospective diary study among adolescents.

The current study examined mechanisms through which trait negative affectivity (NA) and effortful control (EC) prospectively predict increases in depressive symptoms in a community sample of 110 adolescents (mean age=16.40 years) in a 9-week, multi-wave diary study. We hypothesised youth with high NA and low EC would experience greater increases in depressive symptoms because they generate more maladaptive cognitive responses to stress. NA predicted greater event-specific rumination and negative cognitive style across the study period. After controlling for trait cognitive vulnerabilities, only event-specific rumination mediated the effect of NA on subsequent depressive symptoms, suggesting that high-NA youth are more likely to ruminate about stressful events, which then predicts increases in depressive symptoms. EC did not moderate the effects of NA on either depressive symptoms or cognitive responses to stress. Results are discussed in terms of vulnerability-stress models linking temperament and depressive symptoms in adolescence.